Conjugation

Modern Microbial Genetics, Second Edition. Edited by Uldis N. Streips, Ronald E. Yasbin 

Copyright # 2002 Wiley-Liss, Inc. 

ISBNs: 0-471-38665-0 Hardback); 0-471-22197-X Electronic) 

19 

Conjugation 

RONALD D. PORTER 

Department of Biochemistry and Molecular Biology, The Pennsylvania State University, University Park, 

Pennsylvania 16802 

I. Introduction ................................... 464 

A. C. Regulation of F factor fertility. .............. 464 

II. Conjugation by the E. coli FFactor............... 465 

A. Overview ................................... 465 

B. Structure of the F Factor ..................... 465 

C. Regulation of F Factor Fertility ............... 467 

D. Establishment of Cell Contact . . ............... 468 

E. DNA Mobilization and Transfer............... 469 

F. Separation of the Mating Pair . . ............... 471 

III. Hfr Conjugation and Chromosomal Transfer ....... 471 

A. How Hfr Strains Arise . ...................... 471 

B. Properties of Hfr's ........................... 474 

C. Recombination after Hfr conjugation. . . ........ 475 

IV. F-prime Conjugation and Merodiploids ............ 478 

A. The Generation of F-primes................... 478 

B. Conjugation Properties of F-primes ............ 481 

V. Conjugation of Fertility-Inhibited F-like Plasmids . . . 482 

VI. Nonconjugative, Mobilizable Plasmids ............. 483 

VII. Broad Host Range Self-transmissible Plasmids ...... 484 

VIII. Chromosome Mobilization by Non-F Plasmids ..... 486 

IX. Plasmid-Based Conjugation in Other Bacteria....... 488 

A. Salmonella.................................. 488 

B. Pseudomonas................................ 489 

1. Chromosome Mobilization in 

P. aeruginosa ............................. 489 

2. Chromosome Mobilization in P. putida ....... 490 

3. R-primes in Pseudomonads . . ............... 490 

C. Streptomyces................................ 490 

D. Gram-Positive Cocci Streptococcus, etc.) ....... 493 

E. Other Plasmid-Based Conjugation Systems ...... 494 

X. Conjugative Transposons . . ...................... 495 

XI. Conclusion .................................... 496 

XII. Appendix: Conjugational Mapping . ............... 496

464 PORTER 

I. INTRODUCTION 

Conjugation is the mode of gene transfer 

that involves the transfer of DNA between 

two live bacterial cells that are in direct contact. 

Although conjugation in nature most 

often simply involves the transfer of plasmid 

DNA from donor to recipient cell, chromosomal 

DNA can be transferred under certain 

circumstances. Much of the discussion in 

this chapter will focus on the F factormediated 

conjugation system of Escherichia 

coli as this system serves as a prototype for 

conjugation in Gram-negative bacteria. The 

less well-characterized Gram-positive conjugation 

systems will be described later in the 

chapter. 

There are many aspects of the discovery of 

conjugation in E. coli by Lederberg and Tatum 

1946) that were strongly influenced by 

elements of serendipity. The choice of the K- 

12 strain for use in the initial experiments is 

one of the most striking examples of such 

happy chance. The protocol involved mixing 

two isolates that each had at least two nutritional 

deficiencies so that cells where one 

marker had reverted would not be scored as 

recombinants. Although the K-12 strain was 

chosen primarily because of the availability 

of isolates with more than one counterselectable 

nutritional marker, it also happened to 

contain a self-transmissible plasmid. The 

particular self-transmissible plasmid in K- 

12, the F factor, was also unusual in that it 

both constitutively expresses conjugal transfer 

functions see below) and contains several 

transposable elements that allow it to interact 

with the bacterial chromosome. 

A. C. Regulation of F factor fertility 

Lederberg later estimated that only about 

five percent of randomly selected E. coli isolates 

would have given recombinants with the 

selection protocol that was initially employed. 

It was also happy chance that the 

nutritional genetic markers in the isolates 

selected were closely clustered on the 

chromosome in each strain so that optimal 

yields of recombinants were readily obtained 

without the requirement for multiple crossovers. 

Although the work of the Avery 

group with transformation in Streptococcus 

pneumoniae Avery et al., 1944) had set 

the stage, there were still many scientists 

who were reluctant to believe that the lowly 

bacteria could engage in any form of sexual 

activity. The work done by Lederberg and 

others who soon followed convincingly 

demonstrated that bacteria were organisms 

where genetics could be productively practiced. 

There are two distinct requirements that 

must be met in order for conjugation to 

occur. The first of these requirments is that 

the cells be able to engage in a specific contact 

cycle. The second is that some DNA in 

the donor cell be capable of undergoing 

mobilization. Plasmids that encode all of the 

necessary gene products to enable the potential 

donor cell to carry out a specific contact 

cycle with a suitable recipient cell are said 

to be conjugative. Plasmids whose DNA can 

be prepared for transfer to a recipient cell 

are called mobilizable. Both of these capabilities 

do not always reside on the same plasmid, 

however, and neither ability alone 

is sufficient for conjugal DNA transfer. Inability 

to carry out either or both of these 

functions classifies a plasmid as being nonconjugative 

and/or nonmobilizable. A plasmid 

may simply lack one or both of these 

abilities as it was originally isolated, or it may 

have lost one or both of these abilities through 

mutation. Plasmids that are mobilizable, but 

nonconjugative, are often efficiently transferred 

to recipient cells when other plasmids 

present in the donor cell provide the necessary 

cell contact functions. Plasmids that are both 

conjugative and mobilizable are termed selftransmissible 

Clark and Warren, 1979). 

It should be noted, however, that the word 

mobilization is also often used to describe 

the situation where a plasmid, generally a 

self-transmissible one, is able to affect the 

conjugational transfer of donor cell chromosomal 

DNA to a recipient cell. In fact the 

word mobilization is most often used in this 

sense. While plasmid mobilization refers to a

CONJUGATION 465 

plasmid's possession of the ability to transfer 

a copy of its own DNA to a recipient cell 

whenever a mating pair has been formed, 

chromosome mobilization in conjugation 

occurs as the result of some kind of physical 

association between the donor cell chromosome 

and the plasmid undergoing conjugational 

transfer. These plasmid/chromosome 

associations can be very stable in the case 

of Hfr strains or very transient in the case 

of cointegrates formed as intermediates in 

transposition; these examples will be discussed 

later in the chapter. Regardless of the 

degree of stability involved, the chromosome 

is passively carried along to the recipient cell 

as the result of its covalent association with 

the plasmid during chromosomal mobilization. 

The word mobilization will be used 

both ways in this chapter, and the student 

should make every effort not to interchange 

the two meanings of the word. 

II. CONJUGATION BY THE 

E. COLI FFACTOR 

A. Overview 

E. coli cells totally lacking the presence of the 

F factor in any form are called F cells. The 

F factor can, however, exist in a cell in three 

different forms. First, cells containing an autonomously 

replicating F plasmid are called 

F ‡ cells. Such cells efficiently transfer the F 

plasmid to a suitable recipient but rarely 

transfer donor cell chromosomal DNA. 

Second, the F factor is able to integrate 

into the donor cell chromosome to give rise 

to an Hfr high frequency of recombination) 

cell that can efficiently transfer donor cell 

chromosomal DNA to a recipient cell by 

conjugation. Third, F-prime plasmids arise 

when the integrated F factor in an Hfr 

carries some chromosomal DNA with it as 

it is recombined out of the chromosome and 

returns to the autonomously replicating 

state. F-primes are transferred to a suitable 

recipient in much the same manner as a wildtype 

F factor. Although the establishment of 

the mating pair and the initiation of DNA 

transfer is identical in all three cases, the 

ability of these three donor types to transfer 

chromosomal DNA to a recipient cell differs 

considerably. These three different types of F 

factor±containing cells will be discussed in 

more detail later in the chapter. 

In the case of F factor±mediated conjugation, 

contact initially occurs between the tip 

of the donor cell's F factor±encoded sex pilus 

and the exterior envelope of the recipient 

cell. Direct contact, presumably achieved by 

basal disassembly of the pilus, produces an 

unstable mating pair. Multiple cell interactions 

frequently give rise to mating cell 

aggregates that may contain up to 20 cells 

Achtman et al., 1978a). A picture and diagrammatic 

representation of E. coli mating 

aggregates are shown in Figure 1. Although 

some DNA transfer between the cells may 

occur at these early stages, most DNA transfer 

occurs between pairs of cells specifically 

stabilized within the mating aggregate. Cells 

are called exconjugants after mating pair 

dissociation, and recipient cells that have 

received DNA from donor cells are called 

merozygotes. These merozygotes become 

transconjugants after the donor DNA has 

become stabilized in the recipient cell. This 

stabilization of donor DNA can occur either 

by recombination with recipient DNA or, 

in the case of transferred plasmid DNA, by 

establishment of the transferred plasmid 

DNA as an independent replicon in the 

recipient cell. The various steps in this overall 

process will be discussed in more detail 

below. A number of excellent recent review 

articles dealing with F factor±mediated conjugation 

are available for additional study 

Willetts and Skurray, 1980, 1987; Willetts 

and Wilkens, 1984; Ippen-Ihler and Minkley, 

1986, Frost et al., 1994, Firth et al., 1996). 

B. Structure of the F Factor 

The F factor is a 100 kb plasmid that can 

be divided into four fairly distinct regions 

see Fig. 2). The region that is labeled inc, 

rep is the portion of the F factor that is 

involved in the vegetative replication of the 

plasmid. Mini-F plasmids containing only 

this region can be constructed, and these

466 PORTER 

Fig. 1. Hfr times F mating aggregates. An interpretative diagram is shown within each micrograph of mating 

E. coli cells. The Hfr cells in these diagrams are drawn with thin walls while the F cells are drawn with thick 

walls. The cells for which the Hfr versus F assignment was uncertain are shown in white, and F pili that are 

thought to connect cells are indicated. Reproduced from Achtman et al., 1978a, with permission of the 

publisher.) 

mini-F derivatives demonstrate all of the 

replication properties of the parent plasmid. 

It is this region of the plasmid that determines 

the F factor's incompatibility properties 

regarding other plasmids in the same cell 

see Perlin, this volume). 

The F factor also contains a region where 

four transposable elements are clustered. 

In addition to the two copies of IS3 and the 

copy of IS2, there is a copy of Tn1000, 

also known as gd, whose transposition properties 

are very similar to those of Tn3. As


Fig. 2. Map of the Escherichia coli F factor. The four 

major regions of the F factor indicated on the figure 

are 1) the inc,rep region which determines vegetative 

replication and plasmid incompatibility properties, 

2) the tra region which stretches from oriT to 

IS3 and provides conjugative and DNA mobilization 

functions, 3) the region containing the four transposable 

elements that facilitate interactions between 

the F factor and other DNA molecules, and 4) the 

``silent region'' between IS2 and the inc,rep region. 

we will see later in this chapter, it is these 

transposable elements that are primarily responsible 

for the ability of the F factor to 

interact with other DNA molecules, including 

the chromosome, in the cell. The third 

region of the F factor is sometimes referred 

to as the silent region as few distinct 

genetic functions have been shown to reside 

there. 

The approximately 35 kb region of the F 

factor labeled tra is the fourth region, and it 

is involved in making the F factor a selftransmissible 

plasmid. This tra region is 

very similar in organization to the tra regions 

of many F-like R factors. It contains the oriT 

site, at which DNA transfer is initiated, and 

DNA sequence analysis Frost et al., 1994) 

indicates the presence of 36 open reading 

frames with most of the likely genes designated 

tra and some trb. Three of the translated 

genes traM, traJ, and artA) produce 

separate transcripts, but all of the other 

genes form a single operon starting with 

traY. Although this huge operon has secondary 

promoters, the traY promoter appears to 

be dominant under conjugative conditions. 

The overall structure of the tra regulon is 

shown in Figure 3. 

C. Regulation of F Factor Fertility 

The main tra operon starting with traY) is 

positively regulated by the product of the 

separately transcribed traJ gene Willetts, 

1977; Gaffney et al., 1983). A number of 

mechanisms have been proposed to explain 

the need for TraJ protein in the efficient expression 

of the traY promoter, but it is currently 

thought that TraJ protein binding 

works by providing sufficient superhelicity 

for transcription initiation Gaudin and Silverman, 

1993). In most F-like plasmids, the 

traJ gene is normally negatively regulated 

by the finO and finP gene products fin ˆ 

fertility inhibition). A virtue of the F factor 

in genetic studies, however, is its lack of fertility 

inhibition due to its lack of a functional 

finO gene. The tra genes and conjugal ability 

are therefore constitutively expressed unless 

finO is provided in-trans by a fin ‡ F-like plasmid. 

This constitutive expression of fertility 

functions was an important element in the 

Fig. 3. Map of the F factor transfer region. The top line gives a size scale in kilobase pairs, while the second 

line show restriction enzyme sites that are not relevant to our discussion here. The third line represents the 

genes with tra genes shown in uppercase letter and trb genes shown in lowercase letters; the oriT site, the IS3- 

containing finO gene, and the finP transcript are also shown. The remaining lines show the various genes 

grouped by function. Reproduced from Frost et al., 1994, with permission of the publisher.)

468 PORTER 

discovery of conjugation by Lederberg and 

Tatum 1994). The initially mysterious lack 

of fertility regulation in F is due to an IS3 

insertion, which traditionally marks one end 

of the tra regulon, in the finO gene Yoshioka 

et al., 1987). finP is transcribed from the antisense 

strand in the mRNA leader region of 

the traJ gene, but the finP transcript apparently 

does not code for a protein Johnson 

et al., 1981). The overlap of the finP RNA 

and the leader region of the traJ mRNA is 

diagrammatically illustrated in Figure 4. The 

binding of these two RNA molecules places 

the translational start signals for traJ within 

an RNA duplex region which would presumably 

preclude translation initiation Finlay 

et al., 1986). The presence of FinO protein 

has been shown to greatly extend the halflife 

of finP RNA even in the absence of traJ 

mRNA Lee et al., 1992), and the FinO protein 

is capable of binding to secondary structural 

features of both RNA species to 

promote the formation of an RNA duplex 

van Biesen and Frost, 1994; Ghetu et al., 

1999). It is only when the FinO protein stabilizes 

this RNA duplex that the translation of 

the traJ mRNA is precluded. 

Fig. 4. Location of the F factor finP gene. The position 

of the finP transcript is shown relative to the 

traM and traJ genes. The coordinate positions indicated 

are relative to the start of the traJ mRNA. 

The finP transcript starts leftward at position 111 

and extends to roughly position 34. It should be 

noted that the finP transcript is complementary to 

much of the leader region of the traJ mRNA, and 

overlaps slightly with the coding region for the TraJ 

protein. 

The other separately transcribed tra gene, 

traM, is located very close to oriT and appears 

to be negatively autoregulated by means of 

binding sites for its own gene product in the 

dual promoter region Penfold et al., 1996). 

The demonstration that traM is also positively 

regulated by TraY protein binding 

Penfold et al., 1996) has indicated that the 

long suspected possibility of traJ regulation 

of traM is the result of an indirect effect. 

D. Establishment of Cell Contact 

A typical F factor-containing E. coli cell will 

possess one to three F-encoded sex) pili on 

its surface. The F pilus is a hollow cylinder 

with an exterior diameter of 8 nm and an 

interior diameter of 2 nm Folkhard et al., 

1979). These sex pili may be up to 20 mm 

in length, and are often visualized microscopically 

by the adsorption of male-specific 

phages. The F sex pili consist of many molecules 

of a single protein, pilin, which is encoded 

by the traA gene Frost et al., 1984). 

The traQ gene product apparently converts 

the initial 121 amino acid traA polypeptide 

into the functional 70 amino acid polypeptide, 

perhaps while acting as a chaperone 

during its insertion into the inner membrane 

Maneewannakul et al., 1993). The N-terminal 

amino acid of the mature pilin is acetylated 

by the product of the traX gene, but the 

conjugation properties of traX mutants lacking 

this acetylation seem largely unaffected 

Moore et al., 1993). Claims for the phosphorylation 

and glycosylation of pilin have 

not been substantiated. At least 13 additional 

genes are required for the assembly of a functional 

pilus, but their specific roles are not 

known. 

The initiation of mating pair formation 

requires that the tip of the F pilus make contact 

with a specific receptor site on the surface 

of the recipient cell. Although the exact 

nature of that receptor is not known, mutations 

that render a cell incapable of functioning 

as a recipient in conjugation Con ) 

generally map in either ompA or in genes 

involved in lipopolysaccharide LPS) synthesis. 

It appears that LPS participates in the


initiation of mating pair formation, and it has 

also been shown that Zn ‡‡ is required at the 

earliest stages of this process. The expression 

of the ompA gene in the recipient cell, on the 

other hand, is necessary for the stabilization 

of the mating pair. It remains possible, however, 

that the OmpA protein is also part of 

the receptor site, as added LPS-OmpA protein 

complexes block mating pair formation 

more effectively than LPS alone Achtman 

et al., 1978b). The ability of LPS or LPS- 

OmpA protein complexes to prevent mating 

pair formation presumably results from their 

ability to interact with F pili and thereby 

prevent the pili from making contact with 

recipient cells. 

Although DNA transfer can occur through 

an extended F pilus Ou and Anderson, 1970; 

Harrington and Rogerson, 1990), little DNA 

transfer normally occurs before mating pair 

stabilization. The F pilus has been shown to 

retract upon the attachment of male-specific 

phages Jacobson, 1972), and it is generally 

assumed that the initiation of direct envelopeto-envelope 

contact between donor and recipient 

cell involves the retraction of the pilus 

by disassembly at its base within the donor 

cell. The direct contact between cells yields an 

unstable mating pair where little DNA transfer 

occurs, and it is likely that the pilus is 

no longer required after unstable mating 

pair formation has occurred Achtman et al., 

1978a). The conversion of this unstable 

mating pair to a stable mating pair where 

DNA transfer can occur efficiently involves 

the participation of the traG and traN gene 

products, as mutations in either of these genes 

result in inefficient mating pair formation 

without reducing the extent of conjugal 

DNA replication Manning et al., 1981). 

The exact nature of the final surface-to-surface 

interaction between two mating cells is 

not well understood, but the traD gene product 

may be involved in the formation of a 

pore between the two inner membranes Panicker 

and Minkley, 1985). 

Effective mating pair formation between 

two F factor-containing donor cells is prevented 

by a phenomenon called surface exclusion. 

Surface exclusion requires the traS 

and traT gene products which are located in 

the inner and outer membranes, respectively 

Achtman et al., 1977, 1979; Minkley and 

Willetts, 1984; Cheah et al., 1986). Although 

pilus to envelope contacts between donor 

cells do occur, surface exclusion prevents 

mating pair stabilization between two donor 

cells as well as the initiation of donor conjugal 

DNA synthesis see below). Mating between 

two donor cells can be achieved, 

however, by a procedure called F phenocopy 

mating. This involves growing the cell 

to be used as the recipient into stationary 

phase so that tra expression, and therefore 

surface exclusion, is minimized. It is also 

interesting to note that the traT gene product 

becomes a major component of the outer 

membrane and plays a role in serum resistance 

and in reducing the susceptibility of 

cells to phagocytosis Moll et al., 1980; 

Aguero et al., 1984). Although serum resistance 

is not directly relevant to conjugation 

mechanism, this tra-dependent phenotype 

provides a selective advantage to F factorcontaining 

cells in some environments. 

E. DNA Mobilization and Transfer 

For DNA to actually be transferred from 

donor to recipient cell, the plasmid DNA in 

the donor cell must go through a series of 

processing steps that we refer to as mobilization. 

As replacement synthesis of the transferred 

strand of donor cell plasmid DNA is 

generally concurrent with DNA transfer, the 

entire process preparation for transfer, or 

mobilization, and transfer itself) is sometimes 

also referred to as donor conjugal DNA synthesis 

DCDS). Four or five tra gene products 

are involved in DCDS, and several events 

must occur before the actual DNA synthesis 

and transfer begins. First, one strand of the 

DNA is nicked at the F factor oriT origin of 

transfer) site. This nicking was shown by 

infecting cells with a bacteriophage l derivative 

carrying the oriT site, and examining the 

DNA of the progeny phage Everett and Willetts, 

1980). l oriT was used as a convenient 

means of packaging the DNA of interest as it

470 PORTER 

is much easier to isolate the appropriate 

DNA, with and without nicks, from virions 

than to purify a minority species of nicked 

plasmid DNA from cell lysates. It was found 

that 5±10% of the l oriT phage contained a 

nick within oriT when Flac was present in the 

infected cell; there was no nicking observed 

when Flac was not present. 

These and other in vivo studies indicated 

that this strand-specific nicking reaction at 

oriT required the traY gene product plus the 

product of a gene called traZ that later 

turned out to be a secondary translation 

product of the traI gene Everett and Willetts, 

1980; Traxler and Minkley, 1987). Studies 

to determine the precise location of the 

nick at oriT and to carry out the nicking 

reaction in vitro Thompson et al., 1989; 

Matson and Morton, 1991; Reygers et al., 

1991) revealed that the complete traI gene 

product possesses both the oriT nicking activity 

and the strand separation activity 

known as E. coli DNA helicase I Abdel- 

Monem et al., 1983) that is responsible for 

separating the two DNA strands during 

transfer. The traI nickase-helicase apparently 

becomes covalently linked to the 5 0 

end of the nicked DNA strand during the 

nicking reaction and may play a role in plasmid 

recircularization after transfer is complete 

Reygers et al, 1991; Matson et al., 

1993). The traY gene product and the integration 

host factor IHF) of E. coli both 

have binding sites in the oriT region, and 

both are needed for efficient oriT nicking 

by the TraI protein in vitro under more 

physiologically relevant conditions Nelson 

et al., 1995). The nicking reaction at oriT is 

constitutive in that it occurs in the absence of 

either mating pair formation or DCDS 

Everett and Willetts, 1980), so a nick at 

oriT does not automatically lead to the initiation 

of DCDS. It has been suggested that 

the binding of TraY protein and IHF at 

oriT permit the binding of TraI protein to 

form a complex referred to as a relaxosome 

Howard et al, 1995). The presence of such 

relaxosomes is a common attribute of many 

plasmids whose DNA can be self-mobilized. 

The role of the traM gene product has not 

been well defined. It is not required for piliation 

or nicking at oriT, but it is required for 

DNA transfer and replacement strand synthesis 

in the donor cell. It thus seems to 

trigger the start of DCDS at a nicked oriT 

site in response to a signal arising after the 

tip of the F pilus contacts a suitable recipient 

cell Everett and Willetts, 1980). There are 

three TraM protein binding sites near oriT, 

but these are on the 3 0 side of the nick site 

and therefore do not involve the leading end 

of the transferred strand. These TraM binding 

sites clearly play a role in traM autoregulation, 

and the ability of TraM protein to 

also bind TraD protein Disque-Kochem 

and Dreiseikelmann, 1997) may serve to indicate 

a role for TraM protein in positioning 

the nicked DNA at the transfer portal of 

which TraD protein is thought to be a part 

Panicker and Minkley, 1985). 

A single strand of F factor DNA is transferred 

to the recipient cell starting with the 5 0 

end from the nicked oriT site Rupp and 

Ihler, 1968; Ohki and Tomizawa, 1968). 

The transfer of the displaced strand to the 

recipient cell is normally accompanied by 

DNA polymerase III-mediated replacement 

synthesis in the donor cell, but transfer 

can occur in the absence of this synthesis 

Sarathy and Siddiqi, 1973; Kingsman and 

Willetts, 1978). This replacement DNA synthesis 

requires priming by RNA polymerase 

Kingsman and Willetts, 1978), and this requirement 

may reflect blockage of the 3 0 end 

by one of the proteins involved in DCDS. 

A new complementary strand for the 

entering donor DNA is synthesized by the 

recipient cell's normal DNA synthesis machinery. 

It now appears that the necessary 

priming for this complementary strand synthesis 

is achieved by a special promoter 

called Frpo that allows host cell RNA polymerase 

to initiate a transcript on singlestranded 

DNA that can be continued by 

DNA polymerase III holoenzyme Masai 

and Arai, 1997). This special promoter and 

possibly others like it also appear to allow 

for the rapid expression of a number of genes


from the leading region of the transferred F 

factor DNA in the recipient cell before complementary 

DNA strand synthesis has been 

accomplished. While the role of many of 

these leading region genes has not been determined, 

this group includes ssf, the F factor's 

SSB or single-stranded DNA-binding 

protein gene, and the psiB gene whose product 

acts to prevent SOS induction by the 

entering single-stranded DNA in the recipient 

cell during DNA transfer Bailone et al., 

1988; Bagdasarian et al., 1992). 

The stabilization of the F factor in the 

recipient cell is a recA-independent process 

Clark, 1967) that typically requires the 

transfer of both ends of oriT Everett and 

Willetts, 1982). Despite earlier evidence for 

the transfer of single-stranded concatemers 

of F factor DNA Ohki and Tomizawa, 

1968; Matsubara, 1968), the observed requirements 

for recircularization favor the 

transfer of unit length DNA strands Willetts 

and Skurray, 1987). The TraI protein bound 

to the 5 0 end of the transferred strand could 

simply re-ligate that 5 0 end to the 3 0 end when 

it arrives in a direct reversal of the original 

oriT nicking reaction if the 3 0 end remains 

unobstructed during the course of replacement 

DNA synthesis in the donor cell. The 

fact that transfer can occur in the absence of 

replacement DNA synthesis in the donor cell 

indicates that this reaction can most likely 

occur. While the distinct priming event for 

initiation of replacement strand synthesis in 

the donor Kingsman and Willetts, 1978) 

indicates that the 3 0 end is not initially involved 

in rolling circle replication, that result 

does not rule out a later extension event 

involving that 3 0 end. If the free 3 0 end in 

the donor cell is not preserved throughout 

the transfer event, then circularization to 

complete transfer would presumably involve 

another nicking reaction at the reconstituted 

oriT site. While there is no experimental evidence 

that directly supports or contradicts 

this second possibility, the fact that some 

oriT mutations yield reduced nicking efficiency 

without reducing termination or circularization 

efficiency Gao et al., 1994) 

makes it seem unlikely that this is the primary 

mechanism. The student is referred to 

Wilkens and Lanka 1993) for a more extensive 

discussion of this subject. Figure 5 

shows a model for the transfer of F factor 

DNA during conjugation. 

F. Separation of the Mating Pair 

The destabilization of the mating pair and its 

separation are poorly understood. Mechanical 

disruption of the mating pairs leaves 

little apparent lasting damage Low and 

Wood, 1965), and it is therefore possible 

that mating pair disruption is sometimes a 

spontaneous and random process. In Hfr by 

F matings, where the transfer of the tra 

regulon to the recipient occurs only after 

100‡ minutes of DNA transfer see below), 

the mating pairs aggregates) do not show 

detectable levels of separation for at least 

60 minutes Achtman et al., 1978a). When 

F ‡ or F-prime cells are used as donors, however, 

an intact tra regulon is quickly transferred 

to the recipient and mating aggregates 

rapidly decay Achtman et al., 1978a). Although 

there is no firm evidence that initial 

mating pair stability is the same in these two 

situations, it seems reasonable to speculate 

that the expression of transferred tra genes in 

the recipient cell may play an active role in 

mating pair disaggregation. 

III. HFR CONJUGATION AND 

CHROMOSOMAL TRANSFER 

A. How Hfr Strains Arise 

The integration of the F factor into the E. 

coli chromosome gives rise to Hfr strains 

that efficiently transfer, or mobilize, donor 

cell chromosomal markers to recipient cells. 

The first Hfr strains were isolated by Cavalli- 

Sforza HfrC) and Hayes HfrH), and many 

other Hfr strains have subsequently been 

isolated. Although Hfr's representing a minimum 

of 20 clearly distinct sites of F factor 

integration have been described Low, 1972), 

it appears that a limited number of integration 

sites are highly favored. There are 

four transposable elements on the F factor

472 PORTER 

Fig. 5. Model for conjugative transfer of F. A specific strand of the plasmid thick line) is nicked at oriT 

by the TraYI nickase-helicase and transferred in the 5 0 -to-3 0 direction through a pore formed between 

juxtaposed donor and recipient cells. The plasmid strand retained in the donor is shown as a thin line. The 

termini of the transferred strand are attached to the cell membrane. DNA helicase I from the traI gene) is 

bound to the membrane near the pore, and its migration along the transferred strand provides the motive 

force for displacing the strand into the recipient cell. New F factor DNA broken lines) is synthesized in both 

donor and recipient cells by DNA polymerase III. The model assumes that a primer is required for the DNA 

synthesis and that single-stranded DNA-binding protein small circles) coats the single-stranded DNA. 

Reproduced from Willetts and Skurray, 1987, with permission of the publisher.) 

Davidson et al., 1975)Ðtwo copies of IS3, 

one copy of IS2, and one copy of Tn1000 

also known as gdÐand some Hfr formation 

definitely involves recA-dependent recombination 

between an F factor-borne transposable 

element and a homologue in the 

cell's chromosome. It has in fact been shown 

that the sites of Hfr formation largely correlate 

with known positions of IS elements in 

the E. coli chromosome Umeda and Ohtsubo, 

1989). Hfr's arise in recA cells at 

considerably lower frequencies Deonier 

and Mirels, 1977; Cullum and Broda, 1979), 

but the mechanistic basis for this recAindependent 

Hfr formation is not known. 

Hfr's vary greatly in their stability; excision 

of the integrated F factor from some 

chromosomal locations is essentially never 

observed, but F ‡ cells arise at high frequency 

with some Hfr strains Low, 1973). The most 

unstable Hfr's are generally those whose integration 

involved gd, but the basis for the 

variations in stability of other Hfr's is not 

known. The relative position and orientation 

of the integrated F factor for many of the 

commonly used Hfr's is shown in Figure 6. 

The limited ability of an autonomous 

nonintegrated) F factor F ‡ ) to transfer 

the donor chromosome cannot be explained 

solely by the frequency with which Hfr's 

arise in an F ‡ culture. A second component 

of F ‡ -mediated chromosome transfer involves 

a process called conduction. Conduction 

is a type of passive mobilization that 

can involve any replicon, including the 

chromosome, present in an F ‡ cell. When 

the Tn1000 present on the F factor initiates 

replicative transposition see Whittle and


Fig. 6. Map positions of integrated sex factors for some E. coli Hfr strains. Each arrowhead indicates the 

position and orientation of integration of the sex factor on the 100 minute map of the E. coli chromosome. 

The location of some chromosomal genetic markers is also shown. The sequence of markers transferred 

from a given strain begins behind the arrowhead. Thus HfrC located at about 13 minutes) transfers counter 

clockwise from the point of origin purE then lac then argF ) while HfrH located at about 98 minutes) 

transfers clockwise. Reproduced from Low, 1987, with permission of the publisher.) 

Salyers, this volume) to another replicon in 

the donor cell, an intermediate step in the 

transposition process is a cointegrate structure 

where the two copies of Tn1000 serve 

as the boundaries between the F factor and 

the other replicon. Normally such a cointegrate 

structure is rapidly resolved into 

separate replicons in the F ‡ cell by the 

Tn1000-encoded resolvase. The F factor is 

unchanged by this process, but the other 

replicon has a newly added copy of 

Tn1000. If, however, DNA transfer is initiated 

during the cointegrate or replicon 

fusion stage of this transposition, the replicon 

that is covalently linked to the F factor 

by Tn1000 will also be involved in DNA 

transfer. 

When all or part of another replicon is 

transferred to a conjugal recipient by such a 

series of events, we say that it has been ``conducted'' 

by the F factor. This process was 

first described when it was shown that the 

low frequency transfer of pBR322 from 

donor to recipient was always accompanied 

by the addition of a copy of Tn1000 to the 

transferred pBR322 plasmid Guyer, 1978). 

The transposition of Tn1000 from the F 

factor to the chromosome can similarly 

result in the transfer of donor cell chromosomal 

DNA to a recipient cell where it is 

available for recombination with the recipient 

cell chromosome. Although the rapid 

resolution of any such transposition intermediates 

in the donor cell precludes their 

identification as Hfr's, such replicon fusions 

between the F factor and the donor cell 

chromosome temporarily function as Hfr's 

before they are resolved. The Tn1000-based 

conduction of a plasmid by the F factor is 

shown diagrammatically in Figure 7. Other 

transient associations between the F factor 

and the chromosome may also promote

474 PORTER 

Fig. 7. Plasmid conduction. The cell in the upper left-hand corner has a copy of the F factor and a 

nonmobilizable plasmid called pX. The F factor copy of g-d is indicated by the thicker portion of the line 

and the cellular nucleoid is shown as a cross-hatched circle. In step I, g-d transposition to pX is initiated with 

the formation of a cointegrate. Step II shows that this donor cell has formed a mating pair with a suitable 

recipient cell before resolution of the cointegrate, and step III shows transfer of the plasmid cointegrate to 

the recipient cell. In step IV, resolution of the cointegrate occurs independently in both donor and recipient 

cell; pX now has a copy of g-d in both cases. By using a double selection scheme that allows the growth of 

only those recipient cells that express a pX-borne gene typically a drug resistance determinant), a collection 

of g-d insertion mutants can be obtained for any DNA fragment carried by pX. 

chromosome transfer by F ‡ 

Goto et al., 1984). 

donor cells 

B. Properties of Hfr's 

In any typical Hfr strain, the integrated F 

factor resides at a particular location within 

the chromosome, and oriT is pointed in one 

of the two possible directions. This fact leads 

to one of the two most important descriptive 

properties of Hfr conjugation: orientation of 

transfer. The orientation of transfer depends 

on whether oriT is pointed clockwise or 

counterclockwise on the E. coli genetic 

map, and determines the order in which 

chromosomal markers will be transferred 

by the donor. For example, one Hfr strain 

might transfer a set of hypothetical markers 

in the order A then B then C then D, while 

another Hfr strain would transfer D then C 

then B then A.


The second important descriptive property 

of Hfr conjugation is referred to as the 

gradient of transfer. The gradient of transfer 

was originally thought to occur because mating 

pairs undergo spontaneous random disruption 

Jacob and Wollman, 1961), but 

subsequent work has indicated that the time 

dependence of the marker transfer gradient 

is not correlated with the time dependence of 

mating pair disaggregation Wood, 1968; 

Achtman et al., 1978a). Whatever the actual 

mechanism, the net result is that markers 

transferred early are transferred at a higher 

frequency than markers that are transferred 

later. The gradient of transfer dictates that 

marker transfer efficiency will depend on the 

marker's position relative to that of the integrated 

F factor. 

One of the initial questions that arose 

during the characterization of Hfr conjugation 

dealt with the nature of the transfer 

event. The initial data could be explained by 

assuming that a uniformly sized piece of 

donor DNA was always transferred and that 

recombination with the recipient chromosome 

began at the proximal end of the transferred 

donor DNA segment. In that situation 

the gradient of transfer would result from a 

cessation of recombination with time and 

length as processing continued along the fragment. 

It could also be explained by assuming 

that different sized segments of donor DNA 

were transferred and that recombination was 

limited by the size of the piece that had been 

transferred. The latter hypothesis was shown 

to be correct on the basis of experiments involving 

a phenomenon called zygotic induction 

Wollman et al., 1956). 

When an Hfr that is lysogenic for bacteriophage 

l is mated with an F recipient, 

l DNA is transferred to the recipient cell 

without the l cI-encoded repressor. If the 

recipient cell is nonlysogenic, there is no l 

cI-encoded repressor present, and the entering 

l DNA therefore undergoes induction to 

yield a burst of phage in the recipient cell or 

zygote without the need for any recombination. 

This was initially observed in experiments 

where an HfrHl ‡ ) strain was mated 

with an F l ) strain when it was found 

that Gal ‡ transconjugants were essentially 

undetectable. This was the result of the tight 

linkage between the E. coli gal operon and 

the l prophage location on the chromosome; 

rarely were gal genes transferred to the recipient 

without the simultaneous transfer 

of l as the gal operon and the primary bacteriophage 

l integration site are very closely 

linked 0.3 minutes on a 100 minute scale) on 

the E. coli chromosome. At the same time, 

however, recombinants involving markers 

closer to the HfrH point of origin, such 

as Thr ‡ 17minutes earlier) or Lac ‡ 9 minutes 

earlier), were readily detectable in those 

crosses. It was concluded that those transconjugants 

resulted from the transfer of 

shorter pieces of donor DNA that did not 

involve the transfer of the l prophage. 

The nearly uniform rate of transfer of 

DNA from each Hfr makes conjugation a 

powerful tool for genetic mapping over very 

long distances. Transfer is initiated at the 

oriT site of the integrated F factor and proceeds 

in the direction dictated by the orientation 

of its integration. Transfer initiates 

rapidly within about 3 minutes of mixing 

cells) and proceeds at a reasonably uniform 

rate Wood, 1968) of about 45,000 base pairs 

per minute, making time of entry a good 

criterion for determining the distance of a 

marker from the Hfr origin Low, 1987). 

Since transfer is initiated in the middle of 

the integrated F factor, the recipient cells 

remain F Hayes, 1953) unless the entire 

donor chromosome is transferred and the 

F factor is subsequently integrated. Complete 

transfer is rare, but can be detected by 

selecting recombinants for a marker transferred 

late. Chromosomal DNA transferred 

by an Hfr can also recombine with homologous 

plasmid-borne DNA in the recipient 

Porter, 1982). Chromosomal mapping by 

conjugation will be discussed in more detail 

in the Appendix at the end of this chapter. 

C. Recombination after Hfr Conjugation 

Once the transfer of variable portions of the 

Hfr chromosome had been established by

476PORTER 

the zygotic induction experiments described 

above, it became possible to estimate the 

efficiency of recombination events after 

donor DNA transfer. Among recombinants 

selected for a distal marker, more than 50% 

inherit any given nonselected proximal 

marker from the donor. This serves to indicate 

that there is a greater than 50% probability 

of a donor marker being recombined 

into the recipient cell chromosome when 

the presence of a more distal donor marker 

serves to clearly show that the proximal 

marker has been transferred. Inheritance of 

more distal markers in these selected recombinants 

is less frequent, however, probably 

due at least in part to subsequent interruption 

of DNA transfer. 

Very long linkage groups are typically 

observed by genetic criteria in Hfr conjugation. 

One study of marker linkage in Hfr 

conjugation estimated a 20% probability of 

a crossover per ``minute'' one ``minute'' is 

1% of the E. coli chromosomeÐabout 45 

kilobase pairs) of transferred DNA Low, 

1965), while another study estimated an 

even lower frequency of crossovers Pittard 

and Walker, 1967). The net result of these 

long linkage groups is a low frequency of 

recombination between two closely linked 

proximal markers. As an example, you 

might determine the frequency of Thr ‡ , 

Leu ‡ , and Pro ‡ transconjugants among 

those selected for Gal ‡ from a cross between 

HfrH and a multiply marked F recipient. 

Although Gal ‡ transconjugants that are 

plus for all three of these non selected 

proximal markers would be common, recombinants 

that are plus for one or two of 

these markers and minus for the others 

would be considerably less frequent. Although 

you would find that more than 

50% of the Gal ‡ transconjugants were 

``plus'' for any of the three proximal markers 

scored individually, such classes of recombinant 

would show considerable overlap 

for these three markers because the more 

closely linked markers appear to be frequently 

recombined into the recipient chromosome 

as a group. 

Markers very near the Hfr origin, however, 

are not frequently inherited. The rare 

inheritance of very early markers less than 

one to two minutes from the origin) from the 

donor was proposed to be due to a length 

exclusion effect whereby the earliest sequences 

transferred were somehow not available 

for recombination Low, 1965). However, 

crossovers do occur frequently in this very 

early region Pittard and Walker, 1967), suggesting 

that increased crossover frequency 

leads to the reduced recovery of these markers. 

An anti-pairing effect of the leading F 

factor DNA has also been suggested Pittard 

and Walker, 1967). In contrast, the probability 

of crossover per minute of transferred 

DNA is somewhat less for very late markers; 

this effect leads to physically larger linkage 

groups Verhoff and DeHaan, 1966). 

The long linkage groups observed genetically 

are at variance with the results of physical 

studies of recombination following 

conjugation. Differentially labeled donor 

and recipient DNA become covalently associated, 

but only short pieces mostly about 

0.4 kb) of single-stranded donor DNA 

appeared to be integrated Siddiqi and Fox, 

1973). Incorporation of double-stranded 

donor DNA was not detected Siddiqi and 

Fox, 1973), even though the transferred 

single-stranded DNA is rapidly converted 

to the double-stranded state in the recipient 

cell. The method used for detection of 

inserted double-stranded DNA, however, required 

that the light density donor DNA 

initially be found in association with heavy 

density recipient DNA so as to distinguish it 

from unrecombined donor DNA. That criterion 

would be valid if the double-stranded 

insertions were short relative to the broken 

fragments of recipient DNA after cell lysis, 

but it would not be valid for double-stranded 

insertions whose length might be comparable 

to or greater than the recipient DNA fragments 

produced by cell lysis and subsequent 

sample manipulations. It is now accepted 

that large segments of double-stranded 

donor DNA are incorporated into the recipient 

chromosome see below), and the single-


stranded insertions seen by Siddiqi and Fox 

1973) may simply represent heteroduplex 

regions generated by branch migration and 

resolution of Holliday junctions at the crossover 

sites. 

Smith 1991) reevaluated a great deal of 

published linkage data in light of an improved 

understanding of recombination 

mechanism. He suggested that most recombination 

events in E. coli cells with a functional 

RecBCD pathway occur by means 

of RecBCD enzyme entry at both the 

leading end of the transferred DNA and 

the broken end generated by termination 

of transfer. The RecBCD enzyme then processes 

through the DNA in DNA helicase 

mode until it encounters a Chi recombinational 

hotspot an asymmetric 8 base sequence 

that occurs about every 5 kbp in E. 

coli DNA). Nicking at those Chi sites results 

in the displacement of a single-stranded 

DNA tail by continued RecBCD enzyme 

helicase action, and the resulting singlestranded 

tail allows the binding of RecA 

protein for recombination initiation. This 

``long chunk'' mechanism produces a crossover 

at each end of the donor DNA fragment 

so that essentially the entire donor 

sequence is incorporated into the recipient 

cell genome. Smith then proposes that there 

is also a ``short chunk'' mechanism that accounts 

for situations where most of the 

donor DNA sequence proximal to the 

selected marker is not integrated into the 

recipient cell chromosome. The action of 

RecBCD enzyme at the broken or distal 

end of the donor DNA fragment is envisioned 

to be the same in this ``short chunk'' 

case, but the second crossover event does not 

involve RecBCD-dependent recombination 

at the leading end. The speculation is that 

recombination within the transferred donor 

DNA fragment is promoted by the RecF 

recombination pathway by which recombination 

may be initiated by means of the inherent 

partial single-strandedness of recently 

transferred donor DNA. This speculation 

is consistent with published observations 

that strains with an active RecF pathway 

recBCD sbcBC ) show considerably reduced 

linkage following Hfr conjugation as 

compared to recBCD ‡ strains where the 

RecBCD pathway is active and thought to 

predominate Mahajan and Datta, 1979; 

Lloyd and Thomas, 1983). 

Lloyd and Buckman 1995) subsequently 

carried out a study of recombinants formed 

after Hfr conjugation that involved analyzing 

the effect of both distance from the origin of 

transfer and numerous recombination genes. 

Their results were consistent with a mechanism 

such as Smith's RecBCD-dependent 

``long chunk'' model for most recombinants 

where the amount of Hfr donor DNA transferred 

was in the range of 500 kbp or less. 

While Smith's model would require termination 

of DNA transfer to allow entry of the 

RecBCD enzyme at the leading or oriT end, 

they suggest that at least some of these ``long 

chunk'' events may involve non-RecBCDdependent 

initiation events utilizing transient 

single-strandedness near the leading end 

while DNA transfer is still occurring as originally 

proposed by Lloyd and Thomas 

1984). They also observed, however, that 

many of the so-called short chunk recombinants 

where donor markers proximal to the 

selected marker were not incorporated arose 

within sectored colonies that also contained 

recombinants showing the much longer ``long 

chunk'' linkage groups. They propose that 

these sectored colonies may have arisen 

from secondary recombination events involving 

the displaced recipient DNA sequence 

after the recombined donor sequence has 

undergone one round of chromosomal replication. 

As nonsectored short chunk recombinants 

might have arisen from secondary 

recombination events occurring prior to recipient 

cell chromosome replication, they 

regard such secondary recombination events 

are being the probable source for many of the 

short chunk recombinants. 

A somewhat different story emerged when 

the selected marker was such that donor 

DNA segments of 1000 kbp or more had 

to be transferred Lloyd and Buckman, 

1995). While very long linkage groups still

478 PORTER 

predominated under those conditions, it 

appeared that fewer of the proximal or 

leading end crossover events involved donor 

DNA sequence transferred at the earliest 

times. There is no simple, straightforward 

explanation for this phenomenon, but it was 

suggested that recombinants involving longer 

transfer times may more often involve proximal 

end initiation at single-stranded gaps 

that may occur further from the proximal/ 

leading end than RecBCD-dependent events 

occurring after DNA transfer has been terminated. 

The student is referred to the original 

work Lloyd and Buckman, 1995) for a 

discussion of how a number of rec gene 

mutants affect linkage parameters. All of the 

preceding discussion assumes a need for an 

even number of crossover events to produce a 

viable recombinant, and most of that discussion 

has been focused on scenarios involving 

the minimum number of two such events. It 

should be noted, however, that none of the 

data rules out at least the occasional appearance 

of recombinants involving a larger, but 

still even, number of such crossover events. 

When recombination between closely 

linked markers in the lacZ gene was measured 

in transconjugants, it was found that 

there was very little correlation between recombination 

frequency and the map order of 

the alleles as determined by deletion mapping 

Norkin, 1970). This phenomenon lack 

of correlation between recombination frequency 

and physical distance for closely 

linked markers) is referred to as marker 

effects. Such marker effects are thought to 

result from gene conversion events involving 

the nonrandom correction of nucleotide base 

mismatches in heteroduplex DNA produced 

by recombination. The dramatic marker 

effects in Hfr conjugation Norkin, 1970) 

provide a strong argument for the generation 

of heteroduplex DNA during conjugational 

recombination, and this is strengthened by 

the fact that such marker effects subsequent 

to Hfr conjugation have been shown to be 

dependent on the mismatch correction genes 

in E. coli Feinstein and Low, 1986). These 

heteroduplex regions contain one strand 

from each of two different parental DNA 

molecules, and would only occur where a 

single strand of donor DNA became integrated 

into the recipient DNA homoduplex 

perhaps as part of a recombination initiation 

event) or where heteroduplex DNA 

had been generated by branch migration of 

a Holliday junction. As the equivalent of a 

crossover event must involve one of two 

closely linked markers for recombinants to 

be observed, it is not surprising that such 

marker effects are observed. 

Although the recombination that occurs 

after Hfr conjugation is classified as homologous 

or general, there are other considerations 

that may have a bearing on the 

nature and distribution of recombination 

events that occur. fre frequent recombination 

exchange) regions where genetic exchanges 

by the RecF pathway are clustered 

on the E. coli genome have also been suggested 

Bressler et al., 1978, 1981). As already 

discussed above, the location of Chi 

sites may affect the distribution of genetic 

exchanges when the RecBCD pathway is 

involved in conjugational recombination. 

Therefore genetic exchanges between donor 

and recipient DNA are probably not entirely 

random following Hfr conjugation. 

IV. F-PRIME CONJUGATION 

AND MERODIPLOIDS 

A. The Generation of F-primes 

F-prime factors see Low, 1972; Holloway 

and Low, 1987and 1996 for reviews) can 

arise from Hfr's by a number of different 

mechanisms. Those mechanisms include illegitimate 

recombination events those with 

no known mechanistic basis), recombination 

between IS elements that flank the integrated 

F factor, recombination between an IS 

element within an integrated F factor and 

another copy of the same IS element in 

flanking chromosomal sequence, recombination 

between homologous chromosomal sequences 

flanking the integrated F factor, or 

abortive intramolecular transposition where 

the resolution step of Tn1000 transposition


Fig. 8. Formation of F42lac by abortive transposition. The strain in which F42lac arose contained an F 

factor that had integrated at an IS3 between proA,B and lac in the E. coli chromosome. A: The integrated F 

factor, shown as a heavy line, has looped around to bring its copy of g-d into close proximity with the 

chromosome at a point between lac and proC. B: g-d has begun transposition by breakage of the DNA at the 

chromosomal target site and ligation of single strands of g-d at each end of the target site break. C: A 

hypothetical intermediate where the molecules have been realigned and replication of g-d is indicated by the 

arrows. D: Completed transposition event where the replicated copies of g-d have undergone site-specific 

recombination at their internal resolution sites. F42lac formation, however, occurred when the final resolution 

step panel C to panel D) failed to happen. 

does not occur. A schematic representation 

of the abortive intramolecular transposition 

event that gave rise to a particular F-prime 

called F42lac is shown in Figure 8. Early 

studies with F13 a particular F-prime that 

arose in an Hfr13 strain) demonstrated that 

the chromosomal genes present on the F- 

prime were missing from the chromosome 

Scaife and Pekhov, 1964). This observation 

indicates that F-prime formation is mechanistically 

equivalent to a chromosomal deletion 

involving the production of two smaller 

circles from one larger circle in a manner 

that is roughly analogous to prophage excision 

by the Campbell model Low, 1972). 

Such deletion events would normally result 

in the loss of the smaller circle, but it is 

preserved as a newly generated F-prime 

when the deleted sequence includes F factor 

replication functions. Type I F-primes incorporate 

host DNA from only one side of 

the integrated F factor and leave behind part 

of the F factor. Type II F-primes incorporate 

host DNA from both sides of the integrated 

F factor and retain the complete F factor 

see Fig. 9). 

Although the strain in which the F-prime 

arises, the primary F-prime strain, may initially 

contain both an intact and a deleted 

version of the chromosome, the cells remain 

functionally haploid and may require the F- 

prime for viability when only the deleted

480 PORTER 

Fig. 9. Generation of type I and type II F-primes. The insertion of the F factor heavy line with arrow to 

indicate oriT ) within the chromosome thin line) as found in an Hfr is shown at the top of the figure. On the 

left side of the figure, a recombination event between the flanking chromosomal DNA and a site within 

the integrated F factor gives rise to a type I F-prime and leaves some F factor DNA in the chromosome. 

On the right side of the figure, recombination between flanking chromosomal DNA on either side of the 

integrated F factor gives rise to a type II F-prime with the concurrent production of a chromosomal deletion. 

version of the chromosome is present Scaife 

and Pekhov, 1964). Such primary F-prime 

strains do not promote efficient chromosome 

mobilization because the chromosome lacks 

the sequences needed for high-frequency 

homologous recombination with the F- 

prime Pittard and Ramakrishnan, 1964; 

Scaife and Pekhov, 1964; Berg and Curtiss, 

1967). Transfer of an F-prime to a recipient 

with an intact chromosome produces a secondary 

F-prime strain which is merodiploid, 

or partially diploid, for the host chromosomal 

DNA carried by the F-prime. Such 

merodiploids are frequently used for genetic 

complementation experiments, but recA 

mutants should be used to preclude recombination, 

which can lead to confusing results. 

Recombination between the chromosomal 

and F-prime copies of host DNA, and 

subsequent segregation which may not 

be necessary if the recombination event is 

nonreciprocal) can convert an initial heterogenote 

into a homogenote. Such ``homogenotization'' 

is commonly used to move 

alleles between strains Jacob and Wollman, 

1961; Low, 1972). With the use of an appropriate 

selection or screening procedure, an F- 

prime can be used to pick up a mutant allele 

from the chromosome of one strain by 

homogenotization. The F-prime carrying 

that particular mutant allele can then be 

mated into another strain where it is transferred 

to the chromosome by a reversal of 

the homogenotization protocol. After spontaneous 

or acridine orange promoted curing 

of the F-prime Miller, 1972), the second 

strain contains the desired mutant chromosomal 

allele from the first strain with very 

little perturbation of the surrounding chromosomal 

region. The tra-dependent enhanced 

recombination properties of F-primes see 

Section IVB below) cause them to demonstrate 

homogenotization more frequently 

than most other plasmids Yancey and Porter, 

1985). 

Although the isolation of primary F-prime 

strains is not straightforward Berg and Curtiss, 

1967), a number of methods have been 

described for directly obtaining secondary F- 

prime strains Holloway and Low, 1996). 

These methods all involve Hfr times F- 

matings, and the first requirement is an Hfr


strain where the F factor is integrated near 

the chromosomal segment that is to be 

carried by the F-prime being sought. The 

underlying assumption is that F-primes will 

have arisen at a low frequency in any given 

Hfr strain, and that one merely needs a 

means of selecting for the transfer of such 

spontaneously arising F-primes into a suitable 

recipient cell. 

When the marker to be selected for incorporation 

into an F-prime is one that is transferred 

late late-situated) by the Hfr strain 

being used, early interruption of a mating 

often yields F-primes in the selected recipients. 

As the interruption of the mating 

precludes the transfer of a complete donor 

chromosome, the late-situated marker can 

only be transferred to the recipient as part 

of an F-prime. A more general method involves 

the use of recA recipient strains in 

such matings Low, 1968). No recombination 

can occur between donor and recipient 

chromosomes in the recA recipient unless 

recA ‡ is introduced from the donor, and this 

is usually prevented by interruption of the 

mating. Therefore the transferred marker 

can only be selected in the recipient if it is 

part of a functional replicon such as an F- 

prime. Double male strains with two integrated 

copies of the F factor have also been 

used in F-prime isolation Clark et al., 1969). 

When such a strain is mated with a suitable 

recipient, F-primes containing the chromosomal 

region between the two F factor insertion 

sites can be isolated at a relatively high 

frequency. 

Although F-primes involving essentially 

every region of the E. coli chromosome 

have been described Low, 1972), the process 

is far from random and certain F-primes 

repeatedly appear Hadley and Deonier, 

1979, 1980). This preferential formation of 

certain specific F-primes appears to be due 

to the presence and positioning of transposable 

elements Timmons et al., 1983; Umeda 

and Ohtsubo, 1989), and recombination involving 

the multiple copies of ribosomal 

RNA operons rrn) present in the E. coli 

chromosome have also been shown to sometimes 

participate in F-prime formation Blazey 

and Burns, 1983). When transposable 

elements are involved, the necessary circularization 

can occur by general recombination 

between two copies of the transposable element 

or by incomplete replicative transposition 

events initiated by a single copy Hadley and 

Deonier, 1979, 1980; see the chapter by 

Streeps, this volume and Figs. 8 and 9). These 

same types of recombination events may also 

have a major impact on the stability of F- 

primes, particularly large ones. Although F- 

primes containing up to 30% of the E. coli 

chromosome have been reported Low, 

1972), deletion derivatives frequently appear 

in such cultures because the presence of such 

large F-primes slows the growth rate of the 

host cell Simons et al., 1980). It is probable 

that such deletions also frequently occur via 

recombination events involving transposable 

elements. 

B. Conjugation Properties of F-primes 

Due to general recombination in Rec ‡ 

secondary F-prime strains, there is an equilibrium 

between the integrated and autonomous 

state of the F-prime. Because of their 

ability to reintegrate by homologous recombination, 

F-primes were initially described as 

F factors that could ``remember'' where they 

had been integrated Richter, 1957). F- 

primes recombine both into and out of the 

chromosome more frequently than an unaltered 

F factor because the recombination 

events involve extensive stretches of homology; 

dozens to hundreds of kilobases as 

compared to the one to two kilobases of IS 

insertion sequence) homology typically involved 

in F factor integration and excision. 

In its autonomous state the F-prime promotes 

conjugation and self-transfer like the 

F ‡ factor. Such conjugation establishes the 

F-prime in the recipient cell as an independent 

replicon repliconation), but the host 

DNA carried by the F-prime can recombine 

with recipient DNA even when complete repliconation 

does not occur e.g., if part of 

the F DNA is not transferred, as frequently 

happens with large F-primes). The

482 PORTER 

integrated form of the F-prime behaves like 

an Hfr, and thereby serves as the basis of F- 

prime mobilization of the chromosome. This 

is the usage of the word mobilization where a 

plasmid transfers part or all of the donor cell 

chromosome to the recipient cell by means of 

a physical association between plasmid and 

chromosome. F-primes generally mobilize 

the chromosome of a Rec ‡ secondary donor 

cell quite effectively, and transfer efficiencies 

that are only down 5-to 100-fold from Hfrmediated 

chromosome transfer are often 

achieved. F-prime mobilization of the 

chromosome can be used to move specific 

markers by conjugation when a suitable Hfr 

is not readily available. 

The usefulness of F-primes for both 

chromosome mobilization and homogenotization 

is accentuated by their enhanced recombination 

properties. This phenomenon 

was first shown with lplac5 specialized 

transduction where the transducing phage 

recombines with F42lac see Fig. 8) about 

30 times more efficiently than with a 

chromosomal lac gene Porter et al., 1978). 

This recombination enhancement depends 

on both constitutive expression of the F 

factor tra regulon Porter, 1981) and the 

presence of a functional RecBCD enzyme 

exonuclease V) in the recipient cell Porter 

et al., 1978, 1982). Only the oriT site must be 

in-cis on the same DNA molecule as) to the 

recombining lac genes Seifert and Porter, 

1984), and it appears that the action of 

trans-acting tra gene products specifically 

traY and traI) atoriT is crucial in allowing 

enhanced recombination involving F42lac or 

any oriT-containing plasmid Carter and 

Porter, 1991; Carter et al., 1992). The exact 

role of the RecBCD enzyme in promoting 

enhanced recombination is not known, but 

it has been shown that this same recombination 

enhancement phenomenon participates 

in recombination between F42lac and chromosomal 

lac genes in lac merodiploids Yancey 

and Porter, 1985). This tra- and RecBCD 

enzyme-dependent enhancement therefore 

facilitates both chromosome mobilization 

and homogenotization by promoting increased 

recombinational interaction between 

fertile F-primes and the host cell chromosome. 

V. CONJUGATION OF 

FERTILITY-INHIBITED F-LIKE 

PLASMIDS 

Although the conjugation properties of the F 

factor are representative of those seen with a 

variety of self-transmissible plasmids that 

have been examined, it is atypical in that its 

fertility inhibition system is inoperative. As 

mentioned previously, this is because the 

finO gene of the F factor has been inactivated 

by an IS3 insertion. This was fortunate 

for the development of bacterial genetics, but 

it does not appear to be common in nature. 

There are a large number of plasmids, 

however, whose tra regulon structure is similar 

to that of the F factor while their fertility 

inhibition systems remain functional. R 

factor plasmids such as R1 and R100 are 

among the well-characterized members of 

this F-like group. This situation poses two 

questions regarding the general subject of 

fertility inhibition. The first is why fertility 

inhibition is so commonly found in F-like 

plasmids. The second is how significant 

levels of conjugal transfer are achieved under 

fertility inhibition conditions. 

There is no definitive answer to the first 

question, but there is a reasonably good 

speculation that can be advanced. The production 

of pilin and all of the other tra 

proteins that are synthesized when tra expression 

is derepressed must exert a significant 

energy drain on the plasmid-containing 

cell. Although this is of little consequence in 

a laboratory setting, it could represent a significant 

selective disadvantage to a cell that 

is attempting to compete in a natural setting. 

The conferral of serum resistance by the expression 

of the surface exclusion proteins 

could represent a significant selective advantage 

for tra expression in some situations, 

but the surface exclusion genes of the F 

factor are expressed from a secondary promoter 

even when the main tra operon


promoter for traY to traI; see fig. 3) is not 

functioning Cheah et al., 1986). 

The answer to the second question stems 

from the nature of biological regulation. 

Such regulation is essentially never absolute 

as demonstrated by the fact that a low level 

of lacZ expression occurs even when lacI 

repression is functioning. This is also the 

case with the fin system, and a minority of 

the cells in a plasmid-bearing population, are 

always expressing the tra regulon even when 

fertility inhibition is fully functional. This 

minority of conjugally proficient cells is 

capable of initiating conjugal plasmid transfer 

whenever a population of suitable recipient 

cells in encountered. 

The functional donor cells quickly establish 

mating pairs and achieve plasmid transfer 

to a limited number of recipient cells. In 

those transconjugant recipient cells, an expression 

race between fin and tra ensues. 

Initially tra expression wins this race, and 

the primary recipient cells the cells that 

just recieved a copy of the plasmid) readily 

become functional donor cells. These primary 

recipient cells then function as secondary 

donor cells. As this process continues, a 

mating cascade develops that results in 

transfer of the plasmid to a majority of the 

suitable recipient cells that are available. 

Eventually fin expression ``catches up,'' and 

the newly expanded population of plasmidbearing 

cells returns to the state where only a 

small minority of the cells are conjugally 

proficient at any given time. This scenario 

provides a means of increasing genetic variation 

in the overall population without excessive 

energy expenditure by the plasmidbearing 

cells. 

This whole process is analogous to the 

situation observed in the classic PAJAMA 

experiment Pardee et al., 1959). In the 

PAJAMA experiment the name comes from 

PArdee, JAcob, and Monod), there is an expression 

race between the lacI gene and the 

structural genes of the lac operon lacZ, Y, 

and A) when a lacI ‡ lacZ ‡ version of the E. 

coli lac region is transferred to a lacI lacZ 

recipient cell by Hfr conjugation. The burst of 

b-galactosidase synthesis seen in that situation 

indicates that the lacZ gene was expressed 

more rapidly than the lacI gene in 

the recipient cell. Eventually lacI expression 

catches up, and repression of the lac structural 

genes is established. 

The net result in the fin-tra case is a situation 

where a moderate reduction in mating 

proficiency is probably more than compensated 

for by the energy savings experienced 

by the general population. The particular 

subset of cells that will be expressing tra at 

any given time under fertility inhibition conditions 

is random, so no particular cell experiences 

the selective growth disadvantage 

conferred by tra expression for an extended 

period of time. 

VI. NONCONJUGATIVE, 

MOBILIZABLE PLASMIDS 

Many smaller plasmids possess a system that 

frequently provides them with the ability to 

achieve efficient conjugal transfer without 

carrying the considerable number of genes 

that are required for the specific cell contact 

cycle. Whenever these plasmids are present 

in the same cell as an appropriate conjugative 

plasmid, the ability to mobilize their 

DNA allows efficient DNA transfer after 

the formation of a stable mating pair has 

been achieved by the other plasmid. Note 

that here we are using the word mobilization 

in the sense of preparing plasmid DNA for 

transfer, not in the sense of transferring the 

donor cell chromosome. 

While numerous nonconjugative but mobilizable 

plasmids are found both in nature 

and in the scientific literature, the colicin 

E1 ColE1) plasmid see Perlin, this volume) 

will serve as a good example of such a plasmid 

for our purposes here. This 6.8 kb plasmid 

achieves high copy number, and its 

dnaA-independent origin of vegetative replication 

has been used in the construction 

of many of the common E. coli cloning 

vectors. The plasmid contains a nic or bom 

basis of mobility) site that functions much 

like the oriT site of the F factor. Three specific 

proteins were found to be associated

484 PORTER 

with purified ColE1 DNA, and treatment of 

this DNA with ionic detergents or proteases 

caused a relaxation of the superhelical DNA 

to an open circular form Clewell and Helinski, 

1969, 1970). The nick responsible for 

the relaxation occurs at a unique site Lovett 

et al., 1974), and one of the proteins becomes 

covalently associated with the 5 0 end of the 

broken strand during relaxation Guiney 

and Helinski, 1975). It was later inferred 

that the nick introduced during relaxation 

occurs at the nic site, as plasmid derivatives 

with cis-acting mutations incapable of relaxation 

are also incapable of mobilization 

Warren et al., 1978). 

ColE1 also contains a mob mobilization) 

region that contains four genes designated 

mbeA through mbeD where ``mbe'' specifies 

mobilization for ColE1 Boyd et al., 1989). 

Three of these mbe gene products are the 

proteins involved in the relaxation complex, 

and it is the MbeA protein that becomes covalently 

bound to the nicked DNA as is seen 

with TraI protein at the F factor oriT site. The 

role of the fourth Mbe protein is unknown. 

The proteins encoded by these mob the 

generic term) genes are specifically required 

for mobilization, however, as their functions) 

cannot be provided by the conjugative 

plasmid that is responsible for mating pair 

formation. The variation in mob-type gene 

specificities among a number of characterized 

mobilizable and self-transmissible plasmids is 

thought to be the result of differences in the 

oriT-like sites carried by these plasmids Willetts 

and Wilkens, 1984). 

The ability of traI derivatives of the F 

factor to mobilize ColE1 Willetts and 

Maule, 1979) serves to make two additional 

important points. The student will recall see 

above) that the traI gene of the F factor 

encodes the DNA helicase involved in strand 

separation during the transfer of F factor 

DNA Abdel-Monem et al., 1983), and 

traI F factors cannot transfer their DNA 

despite the fact that they can produce stable 

mating pairs. The first point is that the conjugative 

plasmid does not have to be capable 

of transferring its own DNA in order to 

permit transfer of the mobilizable plasmid. 

The second point is that the mobilizable 

plasmid must use a non-F factor host cell 

function or provide its own function for 

strand separation during DNA transfer. 

The complexity of the interactions between 

conjugative and mobilizable plasmids 

is illustrated by the variations in mobilization 

efficiency achieved in the presence of 

different conjugative plasmids. ColE1, for 

example, is efficiently mobilized in the presence 

of IncF, IncI, or IncP conjugative plasmids, 

while mobilization is inefficient in the 

presence of IncW plasmids Reeves and Willetts, 

1974; Warren et al., 1979). RSF1010 is 

another mobilizable plasmid that is efficiently 

mobilized by IncP conjugative plasmids 

while it is not mobilized at all by IncF 

plasmids Willetts and Crowther, 1981). It 

therefore appears that interaction between 

mob-type proteins and an oriT-like site is a 

necessary, but not a sufficient, condition for 

achieving mobilization. A systematic study 

involving all combinations of six distinct 

conjugative plasmids and four mobilizable 

plasmids revealed that interactions between 

the relaxosome of the mobilizable plasmid, 

the traD-like portal or coupling proteins) 

available, and the particular pilus type provided 

by the conjugative plasmid all play an 

important role in mobilization efficiency 

Cabezon et al., 1997). 

VII. BROAD HOST RANGE 

SELF-TRANSMISSIBLE 

PLASMIDS 

While the F factor is capable of conjugal 

transfer to and maintenance in E. coli and 

its close relatives see Section VIIA below), a 

variety of self-transmissible plasmids have a 

much broader host range for both conjugation 

and replication. The best-characterized 

plasmid of this type is known as RK2, RP1, 

and RP4 as the three independently isolated 

plasmids are thought to be essentially identical. 

This plasmid can move between and 

be stably maintained in a large number 

of different Gram-negative bacteria Guiney,


1984). In addition to differences in replication 

competence in different organisms, however, 

plasmids differ in their ability to promote 

mating pair formation and DNA transfer 

across boundaries involving species, genus 

and kingdom. RP4 can promote DNA transfer 

by conjugation to a variety of different 

Gram-positive bacteria Guiney et al., 1985; 

Trieu-Cuot et al., 1987), but selection of 

transconjugants in those experiments requires 

the use of a shuttle plasmid containing 

an origin or replication that is functional in 

the Gram-positive recipients Trieu-Cuot 

et al., 1987). ColE1-derived plasmids can 

also be mobilized from E. coli to filamentous 

cyanobacteria of the genus Anabaena when 

RK2-promoted mating pair formation 

occurs Wolk et al., 1984). Several plasmids 

including the F factor can promote conjugal 

DNA transfer from E. coli to the yeast Saccharomyces 

cerevisiae when the plasmid being 

transferred contains a yeast origin of replication 

Heinemann and Sprague, 1989; Bates 

et al., 1998), but RP4 does this much more 

efficiently than the F factor Bates et al., 

1998). Thus there are also plasmid-dependent 

differences in mating pair formation and 

DNA transfer efficiency that influence such 

DNA transfers across classification boundaries 

as well as in host range for DNA replication. 

The conjugation properties of these broad 

host range plasmids show both similarities 

and differences with the F factor conjugation 

system that are worthy of note. First, RP4 

contains a nic or oriT-like site where plasmid-specific 

proteins act to form a relaxosome 

as part of the DNA mobilization 

process. It also carries the genes involved in 

producing specialized conjugative pili and 

promoting mating pair formation and stabilization. 

However, RK2 encodes a pilus that 

is thinner and more rigid than the F pilus 

Bradley, 1980). One obvious result of this 

difference in pilus structure is that RK2 is 

only capable of mating pair formation on a 

solid surface while the F factor mates equally 

well under either solid or liquid media conditions 

Bradley et al., 1980). Also, the genes 

of the tra regulon of RK2 are not contiguous 

as they are with the F factor in that they are 

divided between two regions called Tra1 and 

Tra2 that constitute almost half of the 

60 kbp plasmid Pansegrau et al., 1994). 

Not all of the genes in these two Tra regions 

are essential for plasmid transfer between 

two E. coli cells those required for such 

transfer are called core genes), and it seems 

possible that at least some of the other genes 

in these regions are involved in transfers involving 

more distantly related organisms. 

Self-transmissible broad host range plasmids 

such as RK2 must overcome several 

barriers in order to cross genus and species 

boundaries. The initial barrier to be overcome 

is stable mating pair formation and 

DNA mobilization. The presence of a restriction 

enzyme system in the recipient cell 

that degrades incoming plasmid DNA potentially 

constitutes a second barrier to successful 

plasmid transfer and establishment. 

As already mentioned, the third possible barrier 

is a replication defect that would preclude 

the autonomous replication of the 

transferred plasmid in the new recipient 

host cell. All of these steps have been shown 

to play a role in determining effective plasmid 

host range. 

One obvious way in which a surface barrier 

could operate would be to prevent the 

donor pilus tip from making suitable contact 

with the surface of a potential recipient cell. 

Purified F sex pili bind to E. coli cells but not 

to Pseudomonas aeruginosa cells Helmuth 

and Achtman, 1978), and it has been shown 

that the F factor cannot promote plasmid 

transfer to P. aeruginosa Guiney, 1982). Experiments 

with a well-characterized series of 

LPS mutants of Salmonella minnesota have 

been used to demonstrate how this surface 

barrier phenomenon can operate. Although 

the progressive reduction in the length of the 

LPS sugar side chain had no effect on the 

ability of the F factor or RK2 to transfer to 

S. minnesota recipients, the ability of R64 an 

IncI plasmid) to transfer dropped off in parallel 

to the reduction in sugar side chain 

length Guiney, 1984). It is also possible

486PORTER 

that future studies of surface phenomena will 

uncover situations where mating pair formation 

and stabilization are inhibited despite 

efficient contact by the tip of the pilus. 

Restriction endonuclease degradation in 

the recipient cell can also potentially limit 

the host range of a self-transmissible plasmid. 

The resident DNA chromosomal and 

plasmid) already in a restriction endonuclease-containing 

cell is resistant to restriction 

because of specific methylation by a 

corresponding modification enzyme, and 

newly replicated cellular DNA is temporarily 

hemimethylated methylated on one strand 

Ðthe parental one) and still protected. Although 

plasmid DNA entering as a singlestrand 

would not be subject to degradation 

by most restriction endonucleases, the 

double-stranded DNA resulting from complementary 

strand synthesis would be unmodified 

and susceptible to degradation. 

One solution for the plasmid would be to 

eliminate possible restriction enzyme recognition 

sites. If RK2 did actually go through a 

process of ``shedding'' restriction enzyme 

recognition sites, the mechanism whereby it 

was accomplished remains a mystery. Digestion 

of RK2 with restriction enzymes recognizing 

a hexanucleotide sequence revealed 

that only one of 18 enzymes tested cut it at 

the expected frequency, while many of these 

enzymes cut this 60 kb plasmid only once or 

twice Lanka et al., 1983). The single EcoRI 

site present in RK2 resulted in a 10-fold 

reduction in its ability to transfer into a 

strain expressing EcoRI, and the introduction 

of several additional EcoRI sites into 

the plasmid caused transfer into an EcoRI 

expressing strain to drop to nearly undetectable 

levels Guiney, 1984). 

An alternative mechanism for dealing with 

restriction activity in the recipient cell is 

exemplified by the ardA gene originally described 

for the ColIb-P9 plasmid and subsequently 

found to be present on a variety of 

self-transmissible plasmids from enteric bacteria 

Chilley and Wilkens, 1995). When expressed 

in the recipient cell, the product of 

this gene interferes with the action of the 

type I restriction endonucleases commonly 

found in enteric bacteria and thus provides 

protection for the entering plasmid. 

Inability of a conjugally transferred plasmid 

to replicate in its new host comprises 

the third major host range barrier. The 

fact, however, that self-transmissible plasmids 

seemingly often promote mating pair 

formation at low efficiency with organisms 

where plasmid replication is not possible, 

can be exploited in experiments utilizing 

nonreplicating ``suicide'' plasmids to introduce 

genetic information into an organism 

by conjugation. The introduction of a transposon 

by this method allows efficient transposon 

mutagenesis of organisms where no 

other method for introducing DNA is available. 

This method can also be used to make 

very specific mutants in such organisms by 

reintroducing cloned fragments of chromosomal 

DNA that have been modified in 

vitro. Recombination with the conjugally 

introduced DNA fragment can result in the 

replacement of the corresponding sequences 

originally present in the chromosome of the 

recipient cell. 

VIII. CHROMOSOME 

MOBILIZATION BY NON-F 

PLASMIDS 

Gram-negative bacteria commonly contain 

plasmids, many of which have conjugative 

or mobilizable properties, or both. Some of 

these plasmids possess a transfer system that 

is highly analogous to that of the F factor 

e.g., overlapping fin specificity or specific 

tra gene complementation), but transfer 

systems quite distinct from F are also found. 

The interrupted tra regulon structure of 

RK2 has already been mentioned Pansegrau 

et al., 1994), and the narrow host range 

IncP plasmid, R91, has also been shown to 

have two separate tra regions Moore and 

Krishnapillai, 1982). tra regulons significantly 

smaller than the 30 ‡ kb regions seen 

with F and F-like plasmids have also been 

described. The entire tra regulon of R46, the 

IncN plasmid on which the SOS-enhancing


muc genes were originally found see DNA 

repair chapter by Yasbin, this volume), has 

been cloned on a 22.5 kb fragment Brown 

and Willetts, 1981), and a 19.4 kb fragment 

from the closely related pCU1 plasmid was 

sufficient to confer self-transmissibility on 

the cloning vector Thatte and Iyer, 1983). 

Tn5 mutagenesis studies with pCU1 have 

suggested that the tra regulon may not even 

occupy the entire 19.4 kb fragment Thatte 

and Iyer, 1983). 

Although most such plasmids rarely affect 

the transfer of donor chromosomal DNA 

to the recipient, there are three possible 

mechanisms by which chromosomal DNA 

mobilization or transfer can theoretically be 

achieved. Here we are using the word mobilization 

to refer to plasmid-mediated transfer 

of donor cell chromosomal DNA; we are not 

using it to describe the preparation of plasmid 

DNA for transfer. First, donor DNA 

may be transferred to a recipient by an Hfrlike 

mechanism after a self-transmissible 

plasmid has stably integrated into the donor 

cell chromosome. Second, donor DNA can 

be transferred by an F-prime-like mechanism. 

If the self-transmissible plasmid carries 

substantial amounts of DNA sequence homologous 

to the donor cell chromosome as 

exemplified by E. coli F-primes), the plasmid 

can recombine in and out of the donor cell 

chromosome. A recA-dependent recombinational 

equilibrium is established between the 

autonomous and integrated states of the 

plasmid, and transfer of donor chromosomal 

DNA is achieved when conjugation initiates 

while the plasmid is in the integrated state. 

Third, the chromosome can be ``conducted'' 

Clark and Warren, 1979) to the recipient 

cell when the plasmid has become involved 

in the formation of a transposition cointegrate 

with the donor cell chromosome. This 

third situation generally represents a transient 

intermediate in replicative transposition 

to or from the plasmid, and is therefore typically 

a low-frequency event. 

Although low-level chromosome mobilization 

often occurs without firm evidence 

for plasmid integration, bonafide Hfr formation 

has only rarely been documented Holloway, 

1979). Hfr-like strains can be derived 

with a number of different plasmids, however, 

by a phenomenon called integrative suppression. 

When strains containing a dnaA ts 

mutation are grown at high temperature, 

various plasmids integrate at low frequency 

into the chromosome at apparently random 

locations to form Hfr-like strains in which 

the integrated plasmid provides the basis 

for chromosomal replication. The DNA 

replication defect of the dnaA ts mutation is 

suppressed, hence the term integrative suppression. 

The F factor, a number of F-like 

plasmids that normally do not integrate into 

the chromosome, and some I-like plasmids 

Datta and Barth, 1976) have all been shown 

to be capable of forming Hfr-like strains by 

integrative suppression. 

Chromosome mobilization by non-F 

plasmids often appears more akin to mobilization 

by F-primes a recombinational 

equilibrium between the autonomous and 

integrated states of a homology bearing plasmid) 

than stable Hfr-like mobilization. The 

same plasmid often gives considerably different 

results in different bacterial species. This 

may often be due to differing degrees and 

amounts of homology; for example, several 

derivatives of plasmid RK2 carrying differing 

amounts of sequence homologous to 

the E. coli chromosome were tested for their 

ability to mobilize the donor cell chromosome. 

The amount of homology between 

the plasmid and the chromosome was directly 

correlated with the ability of the plasmid 

to mobilize the chromosome by a recA ‡ 

-dependent mechanism Grinter, 1981). 

The third mechanism for chromosome 

mobilization is conduction by the formation 

of transposition cointegrates between a selftransmissible 

plasmid and the donor cell 

chromosome. The actual mechanistic basis 

of conduction has been most clearly demonstrated 

for the conduction of pBR322 

a mob , and therefore nonmobilizable, derivative 

of ColE1) to recipient cells by the 

F factor. In this situation, Tn1000 gd) 

forms cointegrates between the F factor and

488 PORTER 

pBR322, and pBR322 transferred by this 

mechanism contains a copy of the transposable 

element left by resolution of the cointegrate 

in the recipient Guyer, 1978; see Fig. 

7). Mutations in genes on nonmobilizable 

plasmids can thus be readily obtained Sancar 

et al., 1981). It is widely assumed that the 

donor cell chromosome can be conducted by 

the same mechanism. 

This mechanism has been shown to be implicated 

in the mobilization actually conduction) 

of the chromosome of a variety of gramnegative 

bacteria by a plasmid called R68.45 

Willetts et al., 1981). The parent R68 plasmid 

another name for RK2) mobilizes the 

chromosome very inefficiently, but the 

R68.45 derivative was isolated on the basis 

of efficient mobilization of the Pseudomonas 

aeruginosa chromosome. R68 contains a 

single copy of IS21 that transposes very infrequently, 

but the two tandem copies of IS21 in 

R68.45 allow very efficient transposition to 

occur. As there was no detectable sequence 

homology between R68.45 and the P. aeruginosa 

chromosome, it was concluded that efficient 

chromosome mobilization was the 

result of high-frequency cointegrate formation 

between the plasmid and the chromosome. 

Other self-transmissible plasmids 

containing a suitable transposable element 

may be able to act similarly. The student is 

referred to a review article by Holloway 

1979) for a detailed discussion of chromosome 

mobilization by plasmids. 

IX. PLASMID-BASED 

CONJUGATION IN OTHER 

BACTERIA 

While much of the effort expended on our 

understanding of conjugation systems has 

involved work with E. coli, conjugational 

phenomena have been described in a wide 

variety of other organisms. Some of these 

additional systems involve conjugative and/ 

or mobilizable plasmids, but conjugative 

transposons have emerged in recent years as 

an increasingly important aspect of conjugation. 

This section will deal with plasmidbased 

conjugation in other bacteria while 

the conjugative transposons will be treated 

separately in the next section. 

A. Salmonella 

Although a variety of natural isolates of 

Salmonella sp. contain self-transmissible 

plasmids, these plasmids have not been extensively 

used in the development of Salmonella 

genetics. Instead, the F factor from E. 

coli K-12 has been imported and utilized to 

develop Hfr strains in S. typhimurium and S. 

abony. Many of the initial F-containing 

strains of Salmonella were largely infertile 

because the strains also contained Fin ‡ plasmids, 

but the curing of those plasmids or the 

use of F factor fertility mutants resulted in 

F-containing Salmonella strains that were 

fully as fertile as their E. coli counterparts 

Sanderson et al., 1983). 

In some strains of S. typhimurium now 

more correctly referred to as S. enterica serovar 

Typhimurium), the F factor integrated 

at only one position on the chromosome, 

and it was postulated that this was due to 

the presence of a sex factor affinity site where 

the F factor could readily recombine at that 

location in the chromosome Sanderson 

et al., 1972). Although the basis for this sex 

affinity site in some strains of S. typhimurium 

is not known, a sex affinity site in the E. coli 

chromosome was shown to be due to the 

presence of gd Guyer et al., 1980). The Hfr 

collection for S. typhimurium and S. abony is 

not as extensive as that for E. coli, but it has 

been used extensively in chromosomal mapping 

in these organisms Sanderson, 1996). 

The power of chromosomal mapping by 

conjugation in Salmonella has been greatly 

extended by the use of a technique employing 

Tn10 insertions positioned around 

the chromosome Chumley et al., 1979). An 

Flac with a temperature-sensitive replication 

defect and containing a copy of Tn10 is 

introduced into a strain carrying Tn10 

at the desired location on the Salmonella 

chromosome. When the temperature is then 

raised, Lac ‡ survivors will result from the 

integration of the Flac by recombination


between the two copies of Tn10. Although 

these pseudo-Hfr's are fairly unstable, a 

fresh isolate functions quite effectively in 

mobilizing donor chromosomal DNA to a 

recipient cell. Variations on this technique 

have also been used in Pseudomonas see 

below), and may well be developed in a 

number of additional organisms. 

B. Pseudomonas 

The pseudomonads are a diverse group of 

microorganisms, and that diversity is reflected 

in the variety of conjugation systems 

that have been developed for use with various 

members of this group. A large number 

of self-transmissible and mobilizable Pseudomonas 

plasmids have been described, and 

some of these have been extensively studied 

because they carry groups of genes involved 

in the metabolism of complex organic compounds 

that bacteria normally cannot metabolize. 

Although conjugation has been 

reported in a number of Pseudomonas species, 

the discussion here will be restricted to 

chromosome mobilization in P. aeruginosa 

and P. putida where conjugation has been 

utilized most extensively for genetic mapping. 

Even with these two species, extensive variations 

in conjugative behavior have been described, 

and plasmids that may mobilize the 

chromosome of one strain are often unable to 

function effectively in other strains of the 

same species. The student is referred to a 

review article Holloway, 1986) for a more 

detailed discussion of conjugation in pseudomonads. 

The student should note that the 

following discussion of chromosome mobilization 

in Pseudomonas deals with plasmidmediated 

transfer of the donor cell chromosome 

rather than the preparation of plasmid 

DNA for transfer. 

1. Chromosome mobilization in P. aeruginosa 

Although a wide variety of chromosome mobilizing 

plasmids have been found in various 

pseudomonads, the beginnings of a defined 

conjugation system in a pseudomonad occurred 

when it was found that a plasmid 

from P. aeruginosa strain PAT could mobilize 

the chromosome of P. aeruginosa strain 

PAO Holloway and Jennings, 1958). The 

plasmid, called FP2, can mobilize the host 

cell chromosome unidirectionally from a 

single chromosomal site. Subsequent usage 

of this plasmid has allowed time-of-entry 

style conjugational mapping see Appendix) 

to be done quite effectively for about 25% of 

the P. aeruginosa chromosome. Other plasmids 

of the FP series with different origin 

sites for chromosome mobilization have 

been described, but their use in mapping 

studies has not been extensive. The mechanism 

by which the FP plasmids mobilize the 

P. aeruginosa chromosome remains unclear. 

Transposon insertions in the chromosome 

have also been used to provide the basis for 

chromosome mobilization in P. aeruginosa. 

This is mechanistically similar to the system 

described for Salmonella above, and involves 

a self-transmissible plasmid containing a 

transposon recombining with another copy 

of the same transposon in the donor cell 

chromosome. R18 and R91±5 are self-transmissible 

plasmids that carry Tn1 in opposite 

orientations relative to the plasmid's origin 

of transfer, and it has been shown that these 

plasmids can mobilize the P. aeruginosa 

chromosome each of the plasmids in the 

opposite direction) starting from a variety 

of chromosomal Tn1 insertions Krishnapillai 

et al., 1981). This methodology contributed 

to the development of a genetic map for 

P. aeruginosa. 

Genetic mapping in P. aeruginosa has also 

been greatly facilitated by the use of a group 

of ECM enhanced chromosome-mobilizing) 

plasmids that is typified by an IncP1 plasmid 

called R68.45 which contains a tandem duplication 

of an insertion sequence called 

IS21. The transposition of IS21 is greatly 

increased when in this tandem duplication 

arrangement, and this allows for reasonably 

frequent conduction of the donor cell 

chromosome when DNA transfer initiates 

while a transposition cointegrate exists. 

R68.45 can mobilize all regions of the 

P. aeruginosa strain PAO chromosome, and 

time-of-entry experiments have been done

490 PORTER 

for the entire chromosome by this methodology 

Haas and Holloway, 1978; Royle 

et al., 1981). 

2. Chromosome mobilization in P. putida 

Although plasmids analogous to the FP 

plasmids of P. aeruginosa have not been 

found in P. putida, a number of genetic 

mapping systems based on chromosomal 

mobilization have been described. The earliest 

system involved chromosome mobilization 

by a hybrid of the Pseudomonas OCT 

and XYL plasmids, and the combining of 

data from this system with transduction 

data allowed the generation of a primitive 

chromosomal map for P. putida PpG1 Mylroie 

et al., 1977). Other conjugational mapping 

work has involved the use of ECM 

plasmids such as R68.45 with P. putida strain 

PPN, but a plasmid called pMO22 has been 

used to develop a much more powerful mapping 

system. 

R91±5 is a Fin derivative of R91 that has 

been used to mobilize the chromosome of 

Tn1containing strains of P. aeruginosa see 

above). This plasmid cannot replicate in P. 

putida, and transconjugants with this organism 

cannot be achieved in matings with P. 

aeruginosa PAO donors. It was found, however, 

that plasmid markers could be recovered 

at a low frequency in P. putida 

recipients when pMO22, an R91±5 derivative 

containing Tn501, was mated from a P. 

aeruginosa PAO donor. Such transconjugants 

turn out to have the entire pMO22 

plasmid integrated into the P. putida 

chromosome, and subsequently function effectively 

as Hfr-like donors in matings with 

other P. putida strains Dean and Morgan, 

1983). A number of distinct chromosomal 

insertion sites for pMO22 were found, and 

the resulting collection of Hfr-like strains has 

been instrumental in the development of a 

genetic map for P. putida strain PPN. Although 

the insertion of either Tn501 or Tn7 

into R91±5 allows such Hfr-like strains to be 

derived, it appears that the Tn1 already present 

in R91±5 provides the means for the 

actual chromosomal insertion event. The 

basis for the coordinate involvement of 

Tn501 or Tn7 with Tn1 in the integration 

process remains a mystery. 

3. R-primes in pseudomonads 

R-primes are derivatives of R factor plasmids 

that have acquired additional DNA 

from the chromosome of a host bacterium. 

These R-primes are largely analogous to the 

F-primes of E. coli discussed above, and the 

extended host range capabilities of many R 

factors has allowed R-primes to be generated 

and utilized in a variety of bacterial genera 

Holloway and Low, 1987, 1996). They have 

been particularly useful in pseudomonads 

where the extensive use of self-transmissible 

R factors has provided a framework for the 

isolation of a variety of R-primes. 

R68.45 and other similar ECM plasmids 

have been extensively used for generating 

R-primes containing P. aeruginosa chromosomal 

DNA. The resulting R-primes contain 

the chromosomal DNA between the two 

copies of IS21 present in the original R68.45 

plasmid. These can be selected for by genetic 

complementation after transfer into suitably 

marked Rec strains of P. aeruginosa, orby 

transfer into P. putida. Genetic complementation 

still occurs in P. putida, but a lack 

of sufficient nucleotide base homology prevents 

recombination between the R-primeborne 

P. aeruginosa DNA and the P. putida 

chromosome. This is analogous to the situation 

that exists when E. coli/S. typhimurium 

merodiploids are made. R-primes can also be 

obtained from strains of P. putida containing 

integrated derivatives of R91±5 by mechanisms 

that appear to be highly analogous 

to F-prime production in Hfr strains of E. 

coli. 

C. Streptomyces 

The importance of members of the genus 

Streptomyces in commercial antibiotic production 

has led to extensive efforts to develop 

genetic systems for these organisms. 

As a result the first conjugative plasmid 

from a Gram-positive organism was found 

in S. coelicolor A32) Vivian, 1971). The


plasmid, SCP1, was characterized on the 

basis of its ability to influence the exchange 

of chromosomal genes and to produce the 

antibiotic methylenomycin. Although it 

does not appear to carry much in the way 

of important genetic information for the cell, 

it is unusual in that it is a linear doublestranded 

DNA plasmid of approximately 

350 kbp Kinashi and Shimaji-Murayama, 

1991). 

The various conjugative behavior patterns 

of the SCP1 plasmid of S. coelicolor are 

somewhat analogous to those of the F factor 

of E. coli. Strains lacking SCP1 SCP1 ; 

originally called UF for ultrafertility) correspond 

to F strains E. coli. Attempted matings 

between two such SCP1 strains that 

also lack all other known fertility plasmids 

yield verifiable recombinants for chromosomal 

markers at extremely low levels Bibb 

and Hopwood, 1981). Such recombinants 

were not detectable in attempted crosses between 

two strains of S. lividans that were 

totally lacking in fertility plasmids Hopwood 

et al., 1983). SCP1 ‡ strains, also called 

IF for initial fertility, appear to correspond 

to F ‡ strains of E. coli in that transfer of the 

presumably autonomous plasmid is several 

orders of magnitude roughly 10 3 to 10 5 

times) more frequent than the detectable 

transfer of chromosomal markers. The fact 

that recombinants from SCP1 ‡ times SCP1 

crosses tend to preferentially inherit chromosomal 

markers from the donor strain has led 

to the conclusion that most of the transfer 

occurs as the result of Hfr-like cells, such as 

NF see below), present in the SCP1 ‡ culture 

Hopwood et al., 1973). 

A variety of Hfr-like derivatives of SCP1 ‡ 

strains of S. coelicolor have been described. 

The original such strain was called NF, for 

normal fertility, and it has been determined 

that the SCP1 plasmid integrates by means 

of recombination between a copy of an insertion 

sequence now called IS446) on the 

plasmid and one of two closely positioned 

copies of IS446 on the S. coelicolor chromosome 

Kendall and Cullum, 1986). The integration 

mechanism also involves the deletion 

of both chromosomal and plasmid sequence, 

but exactly how these deletions arise during 

integration remains unclear. Such strains can 

effectively transfer chromosomal genes to a 

suitable recipient, but the donor properties 

of these strains differ from those of E. coli 

Hfr's in two principal ways Hopwood et al., 

1969). First, the transfer appears to be bidirectional 

in that markers on either side of the 

integrated SCP1 plasmid are transferred 

with equal efficiency. Second, all of the recombinants 

from such NF times UF crosses 

become NF, whereas the vast majority of 

recombinants from an Hfr times F cross 

in E. coli remain F . Although unidirectional 

donors that appear to involve the 

integration of SCP1-prime plasmids see 

below) at alternate chromosomal sites have 

been described, these also involve transfer of 

the donor character to the recipient in all 

cases Vivian and Hopwood, 1973). 

It has been proposed that these two variations 

from Hfr-style chromosomal mobilization 

could be explained by either of 

two models Hopwood et al., 1985). The 

first would have a unidirectional transfer of 

DNA, perhaps single-stranded as in E. coli, 

emanating from the integrated SCP1 with 

the proviso that transfer normally involves 

a DNA length greater than that of the 

chromosome. With this model the unidirectional 

transfer observed in some situations 

might be explained on the basis of a recombination 

hotspot of the SCP1 integration site 

in that particular donor strain. The second 

model is that passive conduction to the recipient 

of the entire chromosome occurs as 

an extension of the fact that the SCP1 plasmid 

itself is transferred to the recipient as an 

intact double-stranded DNA molecule. Although 

experimental evidence to discriminate 

between these two models is not 

available, the second model is made attractive 

by the types of marker transfer gradients 

that are observed in NF times UF crosses. 

Various SCP1-prime plasmids have also 

been described for S. coelicolor Hopwood 

and Wright, 1976). Although the mechanistic 

details of their formation have not been

492 PORTER 

determined, their conjugative properties generally 

correspond to those of F-primes in E. 

coli and R-primes in a variety of Gram-negative 

bacteria. These SCP1-prime plasmids 

can provide for the mobilization of donor 

cell chromosomal markers that are not actually 

present on the SCP1-prime, and both 

stable and unstable Hfr-like strains appear 

to be formed by the integration of these 

plasmids into the chromosome Vivian and 

Hopwood, 1973). 

In addition to SCP1, a variety of other 

fertility plasmids from S. coelicolor have 

been characterized. The first such plasmid 

described was a 31 kb self-transmissible plasmid 

called SCP2 Bibb et al., 1977). This 

plasmid promotes little, if any, transfer of 

chromosomal markers, but variants called 

SCP2* spontaneously arise that are capable 

of low frequency chromosomal transfer. 

Unlike the case with SCP1, neither Hfr-like 

nor F-prime-like donors involving SCP2 

have been found in vivo. The smaller size of 

SCP2, however, has allowed the in vitro construction 

of SCP2-primes that can mediate 

the transfer of the host cell chromosome 


One of the more interesting phenotypes 

associated with Streptomyces conjugation 

is called pocking. These matings typically 

involve the mixing of two strains, often 

starting with spores, on an agar surface, 

followed by mycelial growth, mating, and 

the subsequent transfer of resulting spores 

to selective media to obtain recombinants. 

The regions of the mating plate where plasmid 

transfer has occurred are visible as 

regions of reduced growth or pocksÐit may 

be useful to think of pocks on a Streptomyces 

mating plate as being similar in appearance 

to turbid bacteriophage plaques. 

Plasmid mutants that give rise to small pocks 

also show less extensive plasmid transfer, 

and it has been hypothesized that such small 

pocks arise when the plasmid is incapable of 

intramycelial transfer in the recipient strain, 

referred to as spread, while retaining the ability 

to achieve intermycelial transfer between 

the donor and recipient strains. Although a 

definitive relationship between between plasmid 

transfer and pocking has not been demonstrated, 

the weight of evidence suggests 

that the two phenomena are intimately associated 


A clue as to the basis of pocking appeared 

when it was found that insertions into a 

certain nonreplicative region of pIJ101, a 

small self-transmissible plasmid also used as 

a Streptomyces cloning vector, gave derivatives 

that were incapable of transforming S. 

lividans Kieser et al., 1982). It was originally 

thought that pIJ101 must contain one or 

more kil genes whose expression results in 

cell death unless they are repressed by plasmid-borne 

kor, or kill override, genes Kendall 

and Cohen, 1987). It was thought that 

the original insertion mutant of pIJ101 had 

inactivated a kor gene, and it seemed reasonable 

to conclude that this had made it impossible 

for S. lividans cells containing that 

particular plasmid derivative to survive. 

One might then explain pocking as cell 

growth inhibition by a temporary expression 

of kil genes in recipient cells before establishment 

of kor repression allows cell growth to 

resume. It has not been possible to document 

the presence of bonafide kil genes, however, 

and it is now thought that the kor effect 

stems from its role in transcriptional regulation 

of plasmid genes involved in transfer 

Stein, et al., 1989). The burst of plasmid 

gene expression when this kor repression is 

temporarily alleviated immediately following 

transfer might contribute to both increased 

conjugal proficiency and reduced growth 

rate on the part of new recipient cells. 

Space considerations do not allow a discussion 

of the numerous small self-transmissible 

Streptomyces plasmids that have been 

characterized, but the pIJ101 plasmid serves 

to illustrate a couple of additional important 

features of conjugation in Streptomyces. 

First, only a single tra gene found on pIJ101 

is required for conjugal function on the part 

of the plasmid Kendall and Cohen, 1987), 

so it seems likely that trans-acting host cell 

functions must be involved in the conjugal 

transfer of at least some of the smaller Strep-


tomyces plasmids. Second, a mutational 

analysis of the pIJ101 tra gene has shown 

that plasmid transfer and chromosome mobilizing 

ability are differentially dependent 

on the gene product as some mutations 

affect chromosome mobilizing ability to a 

much greater extent than plasmid transfer 

Pettis and Cohen, 2000). In summary, a 

great deal remains to be learned about 

conjugation mechanism in these interesting 

organisms. For further details, the student 

is referred to Hopwood and Kieser 

1993). 

D. Gram-Positive Cocci Streptococcus, 

etc.) 

Although the streptococci as a group are 

more noted for the transformation systems 

that have been developed in S. pneumoniae 

and S. sanguis, three distinctive types of conjugation 

system have also been described for 

this group of organisms. Two of these systems 

appear to involve plasmid conjugation and 

associated phenomena while the third involves 

a unique type of conjugal transfer of 

chromosomal elements called conjugative 

transposons Section X). One type of plasmid 

conjugation system normally utilizes sex 

pheromones and operates at high efficiency, 

but the other type of plasmid conjugation 

system operates at low efficiency phenomena 

and does not involve the participation of sex 

pheromones. The pheromone-enhanced conjugation 

systems seem largely limited to the 

enterococci see Perlin, this volume). 

Sex pili have not been observed in conjugative 

streptococci, and conjugal transfer 

frequently depends on the co-precipitation 

of donor and recipient cells onto a solid 

surface. High-efficiency mating in liquid culture 

can be obtained, however, with a few 

plasmid-containing Enteroccoccus faecalis 

strains, whose mating is controlled by small 

peptide sex pheromones see Clewell, 1993 

and 1999 for overviews). In this type of plasmid 

conjugation system, the recipient cells 

produce a pheromone that elicits a mating 

response from donor cells. The induction of 

the donor mating response by pheromones 

increases the level of plasmid transfer by 

about 10,000-fold as compared to the level 

observed in the uninduced state Dunny 

et al., 1979; Ike and Clewell, 1984). There 

are a number of these pheromones, and 

each appears to be keyed to a specific plasmid 

Dunny et al., 1979). Genomic analysis 

has recently revealed that a number of the 

common pheromones are processed from the 

signal sequence segments of surface lipoproteins 

Clewell et al., 2000), and a chromosomal 

gene of Enterococcus faecalis called 

eep for enhanced expression of pheromone) 

has been identified that may be involved in 

the specific processing of these signal sequence 

peptides An et al., 1999). 

After receiving the plasmid, the recipient 

cell ceases production of the particular pheromone 

associated with that plasmid, but continues 

to produce other pheromones Dunny 

et al., 1979; Ike et al., 1983). The transconjugant, 

as well as the initial donor cell, also 

produces a new plasmid-encoded peptide 

that functions as a competitive inhibitor of 

the corresponding pheromone. It has been 

speculated that the inhibitor prevents the 

plasmid-containing cells from responding to 

a concentration of pheromone that is too low 

to result in effective mating aggregate formation. 

The nomenclature used for the components 

of these systems is standardized. As an 

example, cells containing plasmid pAD1 respond 

to a pheromone called cAD1 and produce 

an inhibitor that is called iAD1. 

A pheromone-induced donor cell produces 

a surface adhesion protein, referred 

to as the aggregation substance Ehrenfeld 

et al., 1986). This donor aggregation substance 

binds to a surface receptor called the 

enterococcal binding substance to promote 

clumping, and the ability of low concentrations 

of lipoteichoic acid to prevent this 

clumping suggests that it may be a component 

of the binding substance Ehrenfeld et al., 

1986). This enterococcal binding substance is 

present on potential donors and potential 

recipients alike, and pure cultures of donor 

cells exhibit clumping when treated with exogenous 

pheromone.

494 PORTER 

Although successful pheromone-mediated 

plasmid transfer has thus far only been observed 

between E. faecalis cells, one of the E. 

faecalis sex pheromones, called cAM373, 

is also produced by a variety of Staphylococcus 

aureus isolates and by a few isolates of 

other species of streptococci Clewell et al., 

1985). These cAM373-producing isolates 

can induce clumping with E. faecalis strains 

bearing the corresponding plasmid, pAM373, 

but potential recipients containing a functionally 

replicating plasmid were not obtained. 

DNA transfer apparently occurred, 

however, as conjugative transposons see 

Section X) carried by pAM373 were found 

to have successfully integrated into the 

Staphylococcus aureus chromosome after 

such matings. Thus the sex pheromone-mediated 

clumping phenomenon initially described 

for E. faecalis may have a much 

broader role in genetic exchange between 

Gram-positive bacteria. 

Initial genetic analysis with one of these E. 

faecalis plasmids, pAD1, involved the use of 

Tn917 insertions to delineate a 31.2 kb 

region of the plasmid that is necessary for 

high-efficiency plasmid conjugation involving 

pheromones Ehrenfeld and Clewell, 

1987). Insertions in part of this region reduced 

or eliminated plasmid transfer without 

affecting the clumping response, while insertions 

in other regions prevented cell aggregation 

as well as plasmid transfer) in liquid 

cultures. While a number of regulatory genes 

and the gene producing the aggregation substance 

have been studied in some detail for 

several of these plasmids see Clewell, 1999), 

very little is still known about the genes and 

gene products that are directly involved in 

plasmid transfer. Interestingly, the oriT analogue 

for pAD1 is located within a group of 

genes thought to function in the vegetative 

replication of the plasmid An and Clewell, 

1997). 

Considerably less is known about the plasmid 

conjugation system that does not involve 

sex pheromones see Macrina and 

Archer, 1993, for a review). The lack of inducible 

aggregation results in much lower 

plasmid transfer efficiencies, but many of 

these plasmids demonstrate a much broader 

host range. One example is the E. faecalis 

plasmid pAMb1 which transfers to Lactobacillus 

casei, Staphylococcus aureus, and Bacillus 

subtilis, as well as to a variety of other 

species of streptococci. The size of pAMb1, 

approximately 26 kb, is much less than that 

of the conjugative plasmids that are pheromome 

responsive, but the nature of the genes 

whose products mediate the actual transfer 

event has not been determined. Some of 

these conjugative streptococcal plasmids of 

both types pheromone induced and nonpheromone 

induced) can mobilize nonconjugative 

plasmids Dunny and Clewell, 

1975; Smith et al., 1980), and others chromosomal 

DNA Franke et al., 1978). The actual 

basis of mobilization is not yet characterized, 

however, for any of these plasmids. 

E. Other Plasmid-Based Conjugation 

Systems 

It is not our purpose here to catalog all bacterial 

conjugation systems, and many that 

have been described in less well-known organisms 

have not been characterized in any 

detail. The intriguing Ti plasmid conjugation 

system from Agrobacterium is covered separately 

by Ream this volume), and a discussion 

of plasmid conjugation systems in anaerobes, 

particularly the genera Bacteroides and Clostridium, 

may be found in the article by Macrina 

1993) as well as by Whittle and Salyers, 

this volume. We do want to note, however, 

the existence of conjugation in thermophiles 

and hence in the Archaea as well. Since the 

first report of a conjugative plasmid in an 

archaeobacterium of the genus Sulfolobus 

Schleper, et al., 1995), it has been shown 

that conjugative plasmids are quite common 

in that organism Prangishvili et al, 1998; 

Stedman et al. 2000) and that transfer of 

chromosomal markers can be promoted 

Schmidt, et al., 1999). Plasmid conjugation 

has also been demonstrated in a eubacterial 

thermophile of the genus Thermus with the 

additional possibility of Hfr-like transfer suggested 

Ramirez-Arcos, et al., 1998).


X. CONJUGATIVE 

TRANSPOSONS 

These interesting elements can both transpose 

to new locations in the cell in which 

they reside and insert into a replicon in another 

cell after mediating their own transfer 

by conjugation without any stabilized existence 

as a plasmid. They will not be dealt with 

extensively here because they are covered in 

considerable detail by Whittle and Salyers, 

this volume. The student is also referred 

to review articles by Salyers et al. 1995), 

Scott and Churchward 1995), and Rice 

1998). 

The first two of these elements described 

were Tn916 from Enterococcus faecalis 

Franke and Clewell, 1981) and Tn5253, originally 

known as the Vcat-tet) element, 

from Streptococcus pneumoniae Shoemaker 

et al., 1980). The much larger Tn5253 

element turns out to contain a copy of the 

18 kbp Tn916 element, and such composite 

Fig. 10. Proposed model for Tn916 transposition 

and conjugation. a: Staggered nicks in the sequences 

flanking the initial position of the element result in 

the production of nonreplicating circular intermediate 

I with six potentially mismatched bases. While 

this double-stranded circular form may be transferred 

directly to the recipient and then possibly 

undergo correction of the mismatched bases, it is 

perhaps more likely that nicking at the presumptive 

oriT-like site within the element allows transfer of a 

single strand to the recipient cell where complementary 

strand synthesis then occurs; circular intermediate 

II results in either case. b: Cleavage of circular 

intermediate II and the target is followed by ligation 

to produce an integrated element that is flanked by 

mismatched coupling sequences until one round of 

replication has occurred. Reproduced from Scott 

and Churchward, 1995 with permission from the 

`Ànnual Review of Microbiology'' Volume 49 Q 

1995 by Annual Reviews.) 

structures seem to be relatively common 

among conjugative transposons. The intensively 

studied Tn916 element seems to be 

the most common such element among the 

Gram-positive cocci, and many of the other 

elements that have been described are related 

to it. Many of these conjugative transposons 

carry multiple drug-resistance determinants, 

but tetracyline resistance is almost always 

observed to be present in such elements. 

Intercellular transfers by these elements 

only occur at low frequency during filter 

solid surface) matings in the laboratory, 

but interspecies transfer is quite common 

Courvalin, 1994). Tn916 or Tn916-like 

elements presumably of Gram-positive 

origin have been found in a number of 

Gram-negative species, but distinctively different 

conjugative transposons are quite 

commonly found in the Gram-negative 

genus Bacteriodes. 

A model for the excision, conjugation, and 

insertion of Tn916 is shown in Figure 10 

from Scott and Churchward, 1995). An 

element-encoded integrase protein is needed 

for the excision and integration steps while 

the element-encoded excision protein only 

plays a role in the excision step. These proteins 

show some interesting functional similarities 

to the well-known bacteriophage l 

intregase and excision proteins. The fact 

that some strains of Lactococcus lactis can 

serve as a recipient but not a donor for the 

conjugative transposition of Tn916 has led 

to the speculation that an IHF-like host 

factor is also required for excision Bringel 

et al., 1991). The six base pair staggered cuts 

made to initiate excision are called coupling 

sequences, and the excised element is circularized 

even though base-pairing within these 

coupling sequences is unlikely. The integration 

reaction does not duplicate the target 

site as generally seen for more conventional 

transposons, but it does generate flanking 

mismatched heteroduplexes that can be eliminated 

either by mismatch correction or by a 

round of recipient replicon replication. A 

role for the double-stranded circular intermediate 

in the integration reaction is

496PORTER 

supported by the observation that such circular 

intermediates purified from E. coli 

strains can integrate subsequent to protoplast 

transformation in Bacillus subtilis 

Scott et al., 1988). Tn916 does not seem to 

be very site specific with regard to the integration 

reaction, but this is not the case with 

all conjugative transposons. 

The conjugation part of the process is 

much less well understood than the excision 

and integration reactions. One study exploring 

concomitant transfer of chromosomal 

markers suggested the occurrence of something 

approximating a complete cell fusion 

event Torres et al., 1991), but a direct correlation 

of chromosomal marker transfer 

with conjugative transposition was not established. 

At the same time the indirect 

mobilization of donor cell plasmids by conjugative 

transposons in several systems see 

Salyers et al., 1995) suggests that at least 

pore formation between donor and recipient 

cell is mediated by conjugative transposons. 

There is no evidence for either the involvement 

of sex pili as seen with plasmid conjugation 

in the Gram-negative bacteria or the 

involvement of adhesion proteins such as are 

seen with pheromone-mediated plasmid conjugation 

in E. faecalis. It has been suggested 

that a single-strand of the excised doublestranded 

circular element is transferred and 

then duplicated in the recipient cell before 

integration Scott et al., 1994), and oriT-like 

function has been demonstrated for a 466 bp 

segment of Tn916 Jaworski and Clewell, 

1995). Tn5 insertion analysis indicated that 

about 11 kbp located largely towards one 

end of Tn916 consists of genes required for 

intercellular transfer but not intracellular 

transfer of the element Senghas et al., 

1988), and there are 11 putative open reading 

frames in that region. One of those open 

reading frames possesses significant similarity 

to the mbeA gene of ColE1 whose product 

acts at the ColE1 nic site Flannagan 

et al., 1994), but nothing else is known about 

these putative conjugation genes. The traA 

gene of Tn916 is located at the other end of 

the element with the genes for the integase 

and excisase, and traA seems to be a positive 

regulator of other genes involved in the conjugative 

process Jaworski et al., 1996). It 

has been suggested that concerted regulation 

of traA and the int-Tn and xis-Tn genes may 

serve to coordinate the transposition and 

conjugation functions of the element Jaworski 

et al., 1996). 

XI. CONCLUSION 

Bacterial conjugation systems are both diverse 

and complex. In the Gram-negative 

bacteria, conjugation is generally a plasmid 

phenomenon that only occasionally results 

in the transfer of some portion of the donor 

cell chromosome. F factor-mediated transfer 

of the chromosome can be systematically 

achieved with E. coli and some of its close 

relatives, but it is achieved much less frequently 

in a variety of Gram-negative 

bacteria when plasmids other than the F 

factor are utilized. Although conjugation is 

also known to often involve plasmids in 

Gram-positive organisms, the conjugative 

transposons found in both Gram-positive 

and Gram-negative bacteria provide what 

seems to be an exception to the requirement 

for plasmid involvement in all conjugation 

systems. Conjugative donor cell chromosome 

transfer is certainly achieved in the 

streptomycetes, but there are certainly mechanistic 

differences as compared to the E. coli 

F factor system. Nevertheless, this most ascetically 

appealing of the bacterial gene 

transfer processes undoubtedly plays a very 

important role in the production and maintenance 

of genetic variability in bacteria. 

XII. APPENDIX: 

CONJUGATIONAL MAPPING 

A variety of techniques have been developed 

for genetic mapping in bacteria. Both transformation 

and generalized transduction are 

very useful for fine-structure mapping in organisms 

where these gene transfer modes operate 

at reasonably high efficiency. However, 

conjugation using Hfr-type donors remains 

the single most powerful technique for mapping 

large segments of the bacterial chromo-


some. Despite the power of this method, 

conjugational mapping has only been done 

in a handful of organisms where such Hfrtype 

donors have been available. The prototype 

system for such chromosomal mapping 

involves the use of Hfr's based on the E. coli 

F factor. This mapping system has been used 

in some Salmonella species as well as in E. 

coli itself. The principles are the same for the 

Hfr-like systems that have been derived for 

Pseudomonas aeruginosa and P. putida. The 

conjugation system in Streptomyces coelicolor 

has also been used for mapping purposes, 

but the mapping strategy in that organism 

differs considerably from that employed in 

the Gram-negative organisms. The discussion 

here will be limited to the principles 

used in the E. coli-like systems. 

The mapping strategy often used in Hfrstyle 

conjugation is based on a principle 

called time of entry. The rate at which 

DNA is transferred from donor to recipient 

cell is fairly constant for any given E. coli Hfr 

strain, even though some variation between 

Hfr's can be observed. The time after the 

initiation of mating pair formation at which 

any given chromosomal marker enters the 

recipient depends on its position relative to 

the integrated F factor in terms of both distance 

and direction. The E. coli chromosomal 

map is calibrated in minutes to reflect 

this time-of-entry phenomenon. The map 

was originally 90 minutes in length, but it 

was re-calibrated to 100 minutes in the mid- 

1970s in order to more accurately reflect 

numerous mapping data on various parts of 

the chromosome. Map positions given in the 

older literature have to be evaluated in light 

of this re-calibration. 

The actual time-of-entry data obtained reflect 

both the gradient of transfer and orientation 

of transfer that were discussed in the 

chapter. An actual experiment involves the 

following steps. First, the Hfr and F strain 

are grown to a specified cell density generally 

1±2 10 8 cells per ml) at 378C. Temperature 

has a strong effect on fertility F 

factor tra expression is minimal at 308C 

and below) and transfer velocity, so it is 

important to ensure that both strains are 

kept at 378C both prior to and during 

mating. This means that you must adjust 

relative cell density at an early stage of 

growth so that both strains reach the desired 

level at the same time. Second, the two 

strains are mixed, generally at a ratio of 

one Hfr cell per 10 F cells, and incubation 

is continued at 378C. Third, samples of the 

mating culture are taken periodically for 

mating interruption. This is called the blendor 

technique, but a real blendor is seldom 

used. Although vigorous vortex mixing is 

sometimes used to disrupt mating pairs, the 

method of choice involves a special tubeholding 

attachment for a sabre saw. The 

underlying principle is that only so much 

donor DNA transfer can occur in any such 

defined time period. Plating agar is frequently 

present during the interruption 

step, and further initiation of mating is prevented 

by both dilution and cell immobilization 

in the agar after plating. 

The interrupted samples of mating culture 

are plated on the appropriate media for 

scoring the desired recombinants. This involves 

both a selection for the recombinant 

F cells e.g., omitting arginine from minimal 

media to select Arg ‡ recombinants) 

and a counterselection for the Hfr donor. 

Most matings done for mapping purposes 

involve Str s Hfr strains sensitive to the antibiotic 

streptomycin) and Str r recipients so 

that streptomycin can be used as the counterselective 

agent. The data obtained are 

then plotted as recombinants per Hfr or recombinants 

per ml versus minutes of mating 

before blending. Some hypothetical time-ofentry 

data are shown in Figure 11. 

For each genetic marker scored, the timeof-entry 

curve can be broken down into 

three phases as shown for marker A in 

Figure 11. The initial phase is the baseline 

portion of the curve where the scored marker 

has not yet reached any recipients and no 

recombinants can scored. The second phase 

involves the part of the curve where the recombinants 

are beginning to appear and 

the slope of the curve is rapidly changing.

498 PORTER 

Fig. 11. Time-of-entry curve for a hypothetical mating interruption experiment. An Hfr times F mating is 

initiated, and samples of the mating culture are periodically interrupted and plated on plates that select for F 

recipient cells that have received genetic marker A, B, or C from an Hfr donor cell. The time of entry for any 

given marker correlates to its chromosomal location, and the distance of any selected marker from the Hfr 

origin of transfer oriT ) can be determined as described in the text. The three phases of the time-of-entry 

curve for marker A, as described in the text, are shown by numbers. 

The third phase of the curve occurs when the 

slope has stabilized and the number of scorable 

recombinants is rapidly rising. This is 

the part of the curve that is used to determine 

an actual time of entry for a given marker. 

Eventually such curves would show a shoulder 

and plateau as the recombinant level 

maximizes, but this portion of the curve is 

not terribly informative and is generally 

ignored. 

The time of entry for a given marker is 

derived by an extrapolation to the abscissa 

from the earliest part of phase three of the 

curve where recombinants are just beginning 

to appear in significant numbers. The value 

obtained, in minutes of mating before interruption, 

provides a good initial indication of 

the distance between the marker and the site 

of F factor integration. This value is often 

compared to the time of entry achieved in 

similar matings for known markers in that 

region of the chromosome. The use of more 

than one Hfr is frequently used to firm up 

the approximate position before using P1 

transduction to map the marker with precision. 

Although the initiation of DNA transfer is 

not actually synchronous, the rapid rise in 

recombinants at the time-of-entry point often 

closely approximates a step function for 

genetic markers that are transferred early 

by the Hfr. The degree of apparent synchrony 

with which transfer occurs is lessened 

with markers further from the F factor site 

of integration. This is reflected in an increase 

in the phase two region of the curve and 

a reduction in the slope observed in phase 

three of the time-of-entry curve. These effects 

can be seen for markers B and C in Figure 

11. Although useful data can be obtained for


distances up to and beyond 50 minutes, the 

use of an Hfr that transfers the marker of 

interest prior to 30 minutes is preferable for 

mapping purposes. 

All of the discussion to this point has assumed 

that a direct selection is available for 

the marker being transferred to the F recipient. 

When such a direct selection is not 

available, a select-then-screen strategy must 

be employed. Recombinants for an available 

selectable marker are obtained at various 

time points, and these are tested for the nonselectable 

marker by replica plating or some 

other suitable screening procedure. If the 

marker of interest is between the site of F 

factor integration and the selected marker, 

transfer of the marker of interest will be 

observed in roughly half of the recombinants. 

High levels of co-transfer > 75%) indicate 

tight linkage between the selected 

marker and the screened marker. When the 

nonselected marker of interest is transferred 

after the selected marker, co-transfer of the 

two markers will seldom be observed. This 

type of procedure allows the nonselectable 

marker to be assigned to the particular region 

of the chromosome between two such selectable 

markers. The power and ease of drugresistance 

selection has made use of Hfr 

strains with mapped transposon insertions a 

very popular method of carrying out this type 

of select-then-screen protocol. 

When the marker of interest is a conditional 

lethal e.g., auxotrophy or temperature 

sensitivity), a print mating procedure 

can be employed to make the initial assignment 

of marker position to a defined region 

of the chromosome. A plate is prepared with 

about 20 Hfr strains patched on the plate at 

positions corresponding to the appropriate F 

factor integration site on the chromosome. 

The clockwise-transferring Hfr's and counterclockwise-transferring 

Hfr's are generally 

arranged in two concentric circles to simplify 

interpretation of the actual print mating 

plate. A Str r F strain carrying the conditional 

lethal marker to be mapped is grown 

up and spread on a streptomycin-containing 

plate. This plate must either select 

against the conditional lethal marker or be 

incubated under conditions, such as high 

temperature, that provide the necessary selection. 

Then a fresh replica of the Hfr plate 

is printed on the lawn of F cells by replica 

plating. After incubation under suitable conditions, 

this print mating plate is read. For 

each ring of Hfr's, the gradient of transfer 

should be apparent. There will be little or no 

growth at positions where the corresponding 

Hfr is transferring the marker late while 

strong patches will result from Hfr's that 

transfer the marker early. For each set of 

Hfr's clockwise and counterclockwise) a 

sharp breakpoint between early and late 

transfer will be evident. This allows the 

marker to be assigned to the region of the 

chromosome between the F factor integration 

site for the earliest clockwise-transferring 

and the earliest counterclockwisetransferring 

Hfr. Data of this type greatly 

facilitates the choice of Hfr's for subsequent 

interrupted mating experiments. 

When possible, the use of a print mating 

and a subsequent interrupted mating experiment 

provides the most effective means 

of localizing a marker on the E. coli chromosome. 

The availability of numerous Hfr 

strains containing well-mapped transposon 

insertions makes it possible to map essentially 

any genetic marker on the basis of 

time of entry. The student is referred to 

an article by Low 1991) for a more extensive 

discussion of conjugational mapping 

methods. 

LITERATURE CITED 

Abdel-Monem M, Taucher-Scholz G, Klinkert M-Q 

1983): Identification of Escherichia coli DNA helicase 

I as the traI gene product of the F sex factor. 

Proc Natl Acad Sci USA 80:4659±4663. 

Achtman M, Kennedy N, Skurray R 1977): Cell-cell 

interactions in conjugating Escherichia coli: Role of 

traT protein in surface exclusion. Proc Natl Acad Sci 

USA 74:5104 ±5108. 

Achtman M, Morelli G, Schwuchow S 1978a): Cell-cell 

interactions in conjugating Escherichia coli: Role of F 

pili and fate of mating aggregates. J Bacteriol 

135:1053±1061. 

Achtman M, Schwuchow S, Helmuth R, Morelli R, 

Manning G 1978b): Cell-cell interactions in

500 PORTER 

conjugating Escherichia coli: Con mutants and stabilization 

of mating aggregates. Mol Gen Genet 

164:171±183. 

Achtman M, Manning PA, Edelbluth C, Herrlich P 

1979): Export without proteolytic processing of inner 

and outer membrane proteins encoded by F sex factor 

tra cistrons in Escherichia coli minicells. Proc Natl 

Acad Sci USA 76:4837± 4841. 

Aguero CW, Aron L, DeLuca AG, Timmis KN, Cabello 

FC 1984): A plasmid-encoded outer membrane protein, 

TraT, enhances resistance of Escherichia coli to 

phagocytosis. Infect Immun 46:740±746. 

An FY, Clewell DB 1997): The origin of transfer oriT) 

of the enterococcal, pheromone-responding, cytolysin 

plasmid pAD1 is located within the repA determinant. 

Plasmid 37:87±94. 

An FY, Sulavik MC, Clewell DB 1999): Identification 

and characterization of a determinant eep) on the 

Enterococcus faecalis chromosome that is involved in 

the production of the peptide sex pheromone cAD1. J 

Bacteriol 181:5915±5921. 

Avery OT, MacLeod CM, McCarty M 1944): Studies 

on the chemical nature of the substance inducing 

transformation of pneumococcal types: Induction of 

transformation by a desoxyribonucleic acid fraction 

isolated from pneumococcus type III. J Experimental 

Med 79:137±158. 

Bagdasarian M, Bailone A, Angulo JF, Scholz P, Bagdasarian 

MM, Devoret R 1992): PsiB, and anti-SOS 

protein, is transiently expressed by the F sex factor 

during its transmission to an Escherichia coli K-12 

recipient. Mol Microbiol 6:885±893. 

Bailone A, Backman A, Sommer S, Celerier J, Bagdasarian 

MM, Bagdasarian M, Devoret R 1988): PsiB 

polypeptide prevents activation of RecA protein in 

Escherichia coli. Mol Gen Genet 214:389±395. 

Bates S, Cashmore AM, Wilkins BM 1998): IncP plasmids 

are unusually effective in mediating conjugation 

of Escherichia coli and Saccharomyces cerevisiae: Involvement 

of the Tra2 mating system. J Bacteriol 

180:6538±6543. 

Berg CM, Curtiss III R 1967): Transposition derivatives 

of an Hfr strain of Escherichia coli K-12. Genetics 

56:503±525. 

Bibb MI, Freeman RF, Hopwood DA 1977): Physical 

and genetical characterization of a second sex factor, 

SCP2, for Streptomyces coelicolor A32). Mol Gen 

Genet 154:155±166. 

Bibb MI, Hopwood DA 1981): Genetic studies of the 

fertility plasmid SCP2 and its SCP2* variants in 

Streptomyces coelicolor A32). J Gen Microbiol 

126:427± 442. 

Blazey DL, Burns RO 1983): recA-dependent recombination 

between rRNA operons generates type II F 0 

plasmids. J Bacteriol 156:1344 ±1348. 

Boyd AC, Archer JAK, Sherratt DJ 1989): Characterization 

of the ColE1 mobilization region and its protein 

products. Mol Gen Genet 217:488± 498. 

Bradley DE 1980): Morphological and serological relationships 

of conjugative pili. Plasmid 4:155±169. 

Bradley DE, Taylor DE, Cohen DR 1980): Specification 

of surface mating systems among conjugative 

drug resistance plasmids in Escherichia coli K-12. J 

Bacteriol 143:1466±1470. 

Bresler SE, Krivonogov SV, Lanzov VA 1978): Scale of 

the genetic map and genetic control of recombination 

after conjugation in Escherichia coli K-12. Mol Gen 

Genet 166:337±346. 

Bresler SE, Goryshin I-Yu, Lanzov VA 1981): The 

process of general recombination in Escherichia coli 

K-12: structure of intermediate products. Mol Gen 

Genet 183:139±143. 

Bringel F, Van Alstine GL, Scott JR 1991): A host 

factor absent from Lactococcus lactis subspecies lactis 

MG1363 is required for conjugative transposition. 

Mol Microbiol 5:2893±2993. 

Brown AMC, Willetts NS 1981): A physical and genetic 

map of the IncN plasmid R46. Plasmid 5:188±201. 

Cabezon E, Sastre JI, de la Cruz F 1997): Genetic 

evidence of a coupling role for the TraG protein 

family in bacterial conjugation. Mol Gen Genet 

254:400± 406. 

Carter JR, Porter RD 1991): traY and traI are required 

for oriT-dependent enhanced recombination between 

lac-containing plasmids and lplac5. J Bacteriol 

173:1027±1034. 

Carter JR, Patel DR, Porter RD 1992): The role of oriT 

in tra-dependent enhanced recombination between 

mini-F-lac-oriT and lplac5. Genet Res Camb 

59:157±165. 

Cheah K-C, Ray A, Skurray R 1986): Expression of F 

plasmid traT: Independence of traY ! Z promoter 

and traJ control. Plasmid 16:101±107. 

Chilley PM, Wilkens BM 1995): Distribution of the 

ardA family of antirestriction genes on conjugative 

plasmids. Microbiol Reading 141:2157±2164 

Chumley FG, Menzel R, Roth JR 1979): Hfr formation 

directed by Tn10. Genetics 91:639±655. 

Clark AJ 1967): The beginning of a genetic analysis of 

recombination-proficiency. J Cell Physiol 70 Suppl 

1):165±180. 

Clark AJ, Maas WK, Low B 1969): Production of a 

merodiploid strain from a double male strain of E. 

coli K12. Mol Gen Genet 105:1±15. 

Clark AJ, Warren GJ 1979): Conjugal transmission of 

plasmids. Ann Rev Genet 13:99±125. 

Clewell DB 1993): Sex pheromones and the plasmidencoded 

mating response in Enterococcus faecalis. In 

Clewell DB ed): ``Bacterial Conjugation'' New York: 

Plenum Press, pp 349±367. 

Clewell DB 1999): Sex pheromone systems in enterococci. 

In Dunny GM, Winans SC eds): ``Cell-Cell 

Signaling in Bacteria.'' Washington, DC: ASM Press, 

pp 42±65.


Clewell DB, Helinski DR 1969): Supercoiled circular 

DNA-protein complex in Escherichia coli: Purification 

and induced conversion to an open circular 

DNA form. Proc Natl Acad Sci USA 62:1159±1166. 

Clewell DB, Helinski DR 1970): Properties of a supercoiled 

deoxyribonucleic acid-protein relaxation complex 

and strand specificity of the relaxation event. 

Biochem 9:4428± 4440. 

Clewell DB, An FY, White BA, Gawron-Burke C 

1985): Streptococcus faecalis sex pheromome 

cAM373) also produced by Staphylococcus aureus 

and identification of a conjugative transposon 

Tn918). J Bacteriol 162:1212±1220. 

Clewell DB, An FY, Flannagan SE, Antiporta M, 

Dunny GM 2000): Enterococcal sex pheromone precursors 

are part of signal sequences for surface lipoproteins. 


Courvalin P 1994): Transfer of antibiotic resistance 

genes between Gram-positive and Gram-negative bacteria. 

Antimicrob Agents Chemother 38:1447±1451. 

Cullum J, Broda P 1979): Chromosome transfer and 

Hfr formation by F in rec ‡ and recA strains of Escherichia 

coli K-12. Plasmid 2:358±365. 

Datta N, Barth PT 1976): Hfr formation by I pilusdetermining 

plasmids in Escherichia coli K-12. J Bacteriol 

125:811±817. 

Davidson N, Deonier RC, Hu S, Ohtsubo E 1975): 

Electron microscope heteroduplex studies of sequence 

relations among plasmids of Escherichia coli: X. Deoxyribonucleic 

acid sequence organization of F and 

F-primes, and the sequences involved in Hfr formation. 

In Schlessinger D ed): ``Microbiology 1974''. 

Washington, DC: ASM Press, pp 56±65. 

Dean HF, Morgan AF 1983): Integration of R91-5 :: 

Tn501 into the Pseudomonas putida PPN chromosome 

and genetic circularity of the chromosomal map. J 

Bacteriol 153:485± 497. 

Deonier RC, Mirels L 1977): Excision of F plasmid 

sequences by recombination at directly repeated insertion 

sequence 2 elements: involvement of recA. Proc 

Natl Acad Sci USA 74:3965±3969. 

Disque-Kochem C, Dreiseikelmann B 1997): The cytoplasmic 

DNA-binding protein TraM binds to the 

inner membrane protein TraD in vitro. J Bacteriol 

179:6133±6137. 

Dunny GM, Clewell DB 1975): Transmissible toxin 

hemolysin) plasmid in Streptococcus faecalis and its 

mobilization of a noninfectious drug resistance plasmid. 

J Bacteriol 124:784 ±790. 

Dunny GM, Craig RA, Carron RL, Clewell DB 1979): 

Plasmid transfer in Streptococcus faecalis. Production 

of multiple sex pheromones by recipients. Plasmid 

2:454± 465. 

Ehrenfeld EE, Kessler RE, Clewell DB 1986): Identification 

of pheromone-induced surface proteins in 

Streptococcus faecalis and evidence for a role for lipoteichoic 

acid in formation of mating aggregates. J 

Bacteriol 168:6±12. 

Ehrenfeld EE, Clewell DB 1987): Transfer functions of 

the Streptococcus faecalis plasmid pAD1: Organization 

of plasmid DNA encoding response to sex pheromone. 

J Bacteriol 169:3473±3481. 

Everett R, Willetts N 1980): Characterization of an in 

vivo system for nicking at the origin of conjugal DNA 

transfer of the sex factor F. J Mol Biol 136:129±150. 

Everett R, Willetts N 1982): Cloning, mutation, and 

location of the F origin of conjugal transfer. EMBO 

J 1:747±753. 

Feinstein SI, Low KB 1986): Hyper-recombining recipient 

strains in bacterial conjugation. Genetics 

113:13±33. 

Finlay BB, Frost LS, Paranchych W, Willetts NS 1986): 

Nucleotide sequences of five IncF plasmid finP alleles. 

J Bacteriol 167:754±757. 

Firth N, Ippen-Ihler K, Skurray RA 1996): Structure 

and function of the F factor and mechanism of conjugation, 

p.2377±2401. In Neidhardt FC, Curtiss III R, 

Ingraham JL, Lin ECC, Low KB, Magasanik B, Reznikoff 

WS, Riley M, Schaechter M, Umbargar HE 

eds): `Èscherichia coli and Salmonella: Cellular and 

Molecular Biology''. Washington, DC: ASM Press. 

Flannagan SE, Zitzow LA, Su YA, Clewell DB 1994): 

Nucleotide sequence of the 18 kp conjugative transposon 

Tn916 from Enterococcus faecalis. Plasmid 

32:350±354. 

Folkhard W, Leonard KR, Malsey S, Marvin DA, Dubochet 

J, Engel A, Achtman M, Helmuth R 1979): X- 

ray diffraction and electron microscope studies of the 

structure of bacterial F pili. J Mol Biol 130:145±160. 

Franke AE, Dunny GM, Brown BL, An F, Oliver DR, 

Damle SP, Clewell DB 1978): Gene transfer in 

Streptococcus faecalis: Evidence for the mobilization 

of chromosomal determinants by transmissible plasmids, 

p. 45±47. In Schlessinger D ed): ``Microbiology 

1978''. Washington, DC: ASM Press, pp 45±47. 

Franke AE, Clewell DB 1981): Evidence for a chromosome-borne 

resistance transposon Tn916) inStreptococcus 

faecalis that is capable of ``conjugal'' transfer 

in the absence of a conjugative plasmid. J Bacteriol 

145:494±502. 

Frost LS, Paranchych W, Willetts NS 1984): DNA 

sequence of the F traALE region that includes the 

gene for F pilin. J Bacteriol 160:395± 401. 

Frost LS, Ippen-Ihler K, Skurray RA 1994): Analysis 

of the sequence and gene products or the transfer 

region of the F sex factor. Microbiol Rev 58:162±210. 

Gaffney D, Skurray R, Willetts N 1983): Regulation of 

the F conjugation genes studied by hybridization and 

tra-lacZ fusion. J Mol Biol 168:103±122. 

Gao Q, Luo Y, Deonier RC 1994): Initiation and termination 

of DNA transfer at F plasmid oriT. Mol 

Microbiol 11:449± 458. 

Gaudin HM, Silverman PM 1993): Contribution of 

promoter context and structure to regulated expression 

of the F plasmid traY promoter in Escherichia 

coli K-12. Mol Microbiol 8:335±342.

502 PORTER 

Ghetu AF, Gubbins MJ, Oikkawa K, Kay CM, Frost 

LS, Glover JNM 1999): The FinO repressor of bacterial 

conjugation contains two RNA binding regions. 

Biochem 38:14036±14044. 

Goto N, Shoji A, Horiuchi S, Nakaya R 1984): Conduction 

of nonconjugative plasmids by F 0 lac is not 

necessarily associated with transposition of the g-d 

sequence. J Bacteriol 159:590±596. 

Grinter NJ 1981): Analysis of chromosome mobilization 

using hybrids between plasmid RP4 and a fragment 

of bacteriophage l carrying IS1. Plasmid 

5:267±276. 

Guiney DG 1982): Host range of conjugation and replication 

functions of the Escherichia coli sex plasmid 

Flac: Comparison with the broad host range plasmid 

RK2. J Mol Biol 162:699±703. 

Guiney DG 1984): Promiscuous transfer of drug resistance 

in Gram-negative bacteria. J Infect Dis 149:320± 

329. 

Guiney DG, Helinski DR 1975): Association of protein 

with the 5 0 terminus of the broken DNA strand in the 

relaxed complex of plasmid ColE1. J Biol Chem 

250:8796±8803. 

Guiney DG, Chikami G, Deiss C, Yakobson E 1985): 

The origin of plasmid DNA transfer during bacterial 

conjugation. In Helinski DR, Cohen SN, Clewell DB, 

Jackson DA, Hollaender A eds): ``Plasmids in Bacteria''. 

New York: Plenum Press, pp 521±534. 

Guyer MS 1978): The g-d sequence of F is an insertion 

sequence. J Mol Biol 125:233±247. 

Guyer MS, Reed RR, Steitz JA, Low KB 1980): Identification 

of a sex-factor-affinity site in E. coli as g-d. 

Cold Spring Harbor Symp Quant Biol 45:135±140. 

Haas D, Holloway BW 1978): Chromosome mobilization 

by the R plasmid R68.45: A tool in Pseudomonas 

genetics. Mol Gen Genet 158:229±237. 

Hadley RG, Deonier RC 1979): Specificity in formation 

of type II F 0 plasmids. J Bacteriol 139:961±976. 

Hadley RG, Deonier RC 1980): Specificity in the formation 

of Dtra F-prime plasmids. J Bacteriol 

143:680±692. 

Harrington LC, Rogerson AC 1990): The F pilus of 

Escherichia coli appears to support stable DNA transfer 

in the absence of wall-to-wall contact between 

cells. J Bacteriol 172:7263±7264. 

Hayes W 1953): The mechanism of genetic recombination 

in Escherichia coli. Cold Spring Harbor Symp 

Quant Biol 18:75±93. 

Heinemann JA, Sprague Jr GF 1989): Bacterial conjugative 

plasmids mobilize DNA transfer between bacteria 

and yeast. Nature 340:205±209. 

Helmuth R, Achtman M 1978): Cell-cell interactions in 

conjugating Escherichia coli: Purification of F pili 

with biological activity. Proc Natl Acad Sci USA 

75:1237±1241. 

Holloway BW 1979): Plasmids that mobilize the bacterial 

chromosome. Plasmid 2:1±19. 

Holloway BW 1986): Chromosome mobilization and 

genomic organization in Pseudomonas. In Gunsalus 

IC, Sokatch JR, Ornston LN eds): ``The Bacteria: A 

Treatise on Structure and Function'', Vol 10. ``The 

Biology of Pseudomonas''. Orlando, FL: Academic 

Press, pp 217±249. 

Holloway BW, Jennings PA 1958): An infectious fertility 

factor for Pseudomonas aeruginosa. Nature 

181:855±856. 

Holloway B, Low KB 1987): F-prime and R-prime 

factors. In Neidhardt FC, Ingraham JL, Low KB, 

Magasanik B, Schaechter M, Umbargar HE eds): 

`Èscherichia coli and Salmonella typhimurium: Cellular 

and Molecular Biology''. Washington, DC: ASM 

Press, pp 1145±1153. 

Holloway B, Low KB 1996): F-prime and R-prime 

factors. In Neidhardt FC, Curtiss III R, Ingraham 

JL, Lin ECC, Low KB, Magasanik B, Reznikoff 

WS, Riley M, Schaechter M, Umbargar HE eds): 

`Èscherichia coli and Salmonella: Cellular and Molecular 

Biology''. Washington, DC: ASM Press, pp 

2413±2420. 

Hopwood DA, Harold RJ, Vivian A, Ferguson HM 

1969): A new kind of fertility variant in Streptomyces 

coelicolor. Genetics 62:461± 477. 

Hopwood DA, Chater KF, Dowding JE, Vivian A 

1973): Advances in Streptomyces coelicolor genetics. 

Bacteriol Rev 37:371± 405. 

Hopwood DA, Wright HM 1976): Genetic studies on 

SCP1-prime strains of Streptomyces coelicolor A32). 

J Gen Microbiol 95:107±120. 

Hopwood DA, Kieser T, Wright HM, Bibb MJ 1983): 

Plasmids, recombination and chromosome mapping 

in Streptomyces lividans 66. J Gen Microbiol 

129:2257±2269. 

Hopwood DA, Lydiate DJ, Malpartida F, Wright HM 

1985): Conjugative sex plasmids of Streptomyces. In 

Helinski DR, Cohen SN, Clewell DB, Jackson DA, 

Hollaender A eds): ``Plasmids in Bacteria''. New 

York: Plenum Press, pp 615±634. 

Hopwood DA, Kieser T 1993): Conjugative plasmids of 

Streptomyces. In Clewell DB eds): ``Bacterial Conjugation''. 


Howard MT, Nelson WC, Matson SW 1995): Stepwise 

assembly of a relaxosome at the F plasmid origin of 

transfer. J Biol Chem 270:28381±28386. 

Ike Y, Craig RC, White BA, Yagi Y, Clewell DB 1983): 

Modification of Streptococcus faecalis sex pheromones 

after acquisition of plasmid DNA. Proc Natl 

Acad Sci USA 80:5369±5373. 

Ike Y, Clewell DB 1984): Genetic analysis of the pAD1 

pheromone response in Streptococcus faecalis, using 

transposon Tn917 as an insertional mutagen. J Bacteriol 

158:777±783. 

Ippen-Ihler K, Minkley Jr EG 1986): The conjugation 

system of F, the fertility factor of Escherichia coli. 

Ann Rev Genet 20:593±624.


Jacob F, Wollman EL 1961): Sexuality and the genetics 

of bacteria. New York: Academic Press. 

Jacobson A 1972): Role of F pili in the penetration of 

bacteriophage f1. J Virol 10:835±843. 

Jaworski DD, Clewell DB 1995): A functional origin of 

transfer oriT) on the conjugative transposon Tn916.J 

Bacteriol 177:6644 ±6651. 

Jaworski DD, Flannagan SE, Clewell DB 1996): Analysis 

of traA, int-Tn, and xis-Tn mutations in the 

conjugative transposon Tn916. Plasmid 36:201±208 

Johnson D, Everett R, Willetts N 1981): Cloning of F 

DNA fragments carrying the origin of transfer oriT 

and the fertility inhibition gene finP. J Mol Biol 

153:187±202. 

Kendall K, Cullum J 1986): Identification of a DNA 

sequence associated with plasmid integration in Streptomyces 

coelicolor A32). Mol Gen Genet 

202:240±245. 

Kendall KJ, Cohen SN 1987): Plasmid transfer in 

Streptomyces lividans: Identification of a kil-kor 

system associated with the transfer region of pIJ101. 

J Bacteriol 169:4177±4183. 

Kieser T, Hopwood DA, Wright HM, Thompson CJ 

1982): pIJ101, a multi-copy broad host-range Streptomyces 

plasmid: Functional analysis and development 

of DNA cloning vectors. Mol Gen Genet 

185:223±238. 

Kinashi H, Shimaji-Murayama M 1991): Physical characteristics 

of SCP1, a giant linear plasmid from Streptomyces 

coelicolor. J Bacteriol 173:5123±5129. 

Kingsman A, Willetts N 1978): The requirements for 

conjugal DNA synthesis in the donor strain during 

Flac transfer. J Mol Biol 122:287±300. 

Krishnapillai VN, Royle P, Lehrer J 1981): Insertions 

of the transposon Tn1 into the Pseudomonas aeruginosa 

chromosome. Genetics 97:495±511. 

Lanka E, Lurz R, Furste JP 1983): Molecular cloning 

and mapping of SphI restriction fragments of plasmid 

RP4. Plasmid 10:303±307. 

Lederberg J, Tatum EL 1946): Gene recombination in 

Escherichia coli. Nature 158:558. 

Lee SH, Frost LS, Paranchych W 1992): FinOP repression 

of the plasmid involves extension of the half-life 

of FinP antisense RNA by FinO. Mol Gen Genet 

235:131±139. 

Lloyd RG, Thomas A 1983): On the nature of the 

RecBC and RecF pathways of conjugal recombination 

in Escherichia coli. Mol Gen Genet 190:156±161. 

Lloyd RG, Thomas A 1984): A molecular model for 

conjugational recombination in Escherichia coli K12. 

Mol Gen Genet 197:328±336. 

Lloyd RG, Buckman C 1995): Conjugational recombination 

in Escherichia coli: Genetic analysis of recombinant 

formation in Hfr x F-crosses. Genetics 

139:1123±1148. 

Lovett MA, Guiney DG, Helinski DR 1974): Relaxation 

complexes of plasmids ColE1 and ColE2: 

Unique site of the nick in the open circular DNA of 

the relaxed complexes. Proc Natl Acad Sci USA 

71:3854±3857. 

Low B 1965): Low recombination frequency for 

markers very near the origin in conjugation in E. 

coli. Genet Res 6:469± 473. 

Low B 1968): Formation of merodiploids in matings 

with a class of Rec recipient strains of Escherichia 

coli K12. Proc Natl Acad Sci USA 60:160±167. 

Low KB 1972): Escherichia coli K-12 F-prime factors, 

old and new. Bacteriol Rev 36:587±607. 

Low B 1973): Rapid mapping of conditional and auxotrophic 

mutations in Escherichia coli K-12. J Bacteriol 

113:798±812. 

Low KB 1987): Mapping techniques and determination 

of chromosome size. In Neidhardt FC, Ingraham JL, 

Low KB, Magasanik B, Schaechter M, Umbargar HE 

eds): `Èscherichia coli and Salmonella typhimurium: 

Cellular and Molecular Biology''. Washington, DC: 

ASM Press, pp 1184±1189. 

Low KB 1991): Conjugational methods for mapping 

with Hfr and F-prime strains. Methods Enzymol 

204:43±62. 

Low B, Wood TH 1965): A quick and efficient method 

for interruption of bacterial conjugation. Genet Res 

6:300±303. 

Macrina FL 1993): Conjugal transfer in anaerobic bacteria. 

In Clewell DB ed): ``Bacterial Conjugation''. 


Macrina FL, Archer GL 1993): Conjugation and broad 

host range plasmids in streptococci and staphylococci. 

In Clewell DB ed): ``Bacterial Conjugation''. New 

York: Plenum Press, pp 313±329. 

Mahajan SK, Datta AR 1979): Mechanisms of recombination 

by the RecBC and the RecF pathways 

following conjugation in Escherichia coli. Mol Gen 

Genet 169:67±78. 

Maneewannakul K, Maneewannakul S, Ippen-Ihler K 

1993): Synthesis of F pilin. J Bacteriol 175:1384 ± 

1391. 

Manning PA, Morelli G, Achtman M 1981): traG protein 

of the F sex factor of Escherichia coli and its role 

in conjugation. Proc Natl Acad Sci USA 

78:7487±7491. 

Masai H, Arai K 1997): Frpo: A novel single-stranded 

DNA promoter for transcription and for primer RNA 

synthesis of DNA replication. Cell 89:897±907. 

Matson SW, Morton BS 1991): Escherichia coli DNA 

helicase I catalyzes a site-and strand-specific nicking 

reaction at the F plasmid oriT. J Biol Chem 

266:16232±16237. 

Matson SW, Nelson WC, Morton BS 1993): Characterization 

of the reaction product of the oriT nicking 

reaction catalyzed by Escherichia coli DNA helicase I. 

J Bacteriol 175:2599±2606.

504 PORTER 

Matsubara K 1968): Properties of sex factor and related 

episomes isolated from purified Escherichia coli 

zygote cells. J Mol Biol 38:89±108. 

Miller JH 1972): `Èxperiments in Molecular Genetics.'' 

Cold Spring Harbor, NY: Cold Spring Harbor Laboratory 

Press. 

Minkley EG Jr, Willetts NS 1984): Overproduction, 

purification, and characterization of the F traT protein. 


Moll A, Manning PA, Timmis K 1980): Plasmid-determined 

resistance to serum bactericidal activity: A 

major outer membrane protein, the traT gene product 

is responsible for plasmid-specified serum resistance 

in Escherichia coli. Infect Immun 28:359±367. 

Moore D, Hamilton CM, Maneewannakul K, Mintz Y, 

Frost LS, Ippen-Ihler K 1993): The Escherichia coli 

K-12 F plasmid gene traX is required for acetylation 

of F pilin. J Bacteriol 175:1375±1383. 

Moore RJ, Krishnapillai V 1982): Tn7 and Tn501 insertions 

into Pseudomonas aeruginosa plasmid R91-5: 

mapping of two transfer regions. J Bacteriol 

149:276±283. 

Mylroie JR, Friello DA, Siemens TV, Chakrabarty AM 

1977): Mapping of Pseudomonas putida chromosomal 

genes with a recombinant sex factor plasmid. 


Nelson WC, Howard MT, Sherman JA, Matson SW 

1995): The traY gene product and Integration Host 

Factor stimulate Escherichia coli DNA helicase I- 

catalyzed nicking at the F plasmid oriT. J Biol Chem 

270:28374 ±28380. 

Norkin LC 1970): Marker-specific effects in genetic 

recombination. J Mol Biol 51:633±655. 

Ohki M, Tomizawa J 1968): Asymetric transfer of 

DNA strands in bacterial conjugation. Cold Spring 

Harbor Symp Quant Biol 33:651±657. 

Ou JT, Anderson TF 1970): Role of pili in bacterial 

conjugation. J Bacteriol 102:648±654. 

Panicker MM, Minkley EG Jr 1985): DNA transfer 

occurs during a cell surface contact stage of F sex 

factor-mediated bacterial conjugation. J Bacteriol 

162:584±590. 

Pansegrau W, Lanka E, Barth PT, Figurski DH, Guiney 

DG, Haas D, Helinski DR, Schwab H, Stanisich VA, 

Thomas CM 1994): Complete nucleotide sequence of 

Birmingham IncPa plasmids: Compilation and comparative 

analysis. J Mol Biol 239:623±663. 

Pardee AB, Jacob F, Monod J 1959): The genetic control 

and cytoplasmic expresssion of inducibility in the 

synthesis of b-galactosidase by E. coli. J Mol Biol 

1:165±178. 

Penfold SS, Simon J, Frost LS 1996): Regulation of the 

expression of the traM gene of the F sex factor of 

Escherichia coli. Mol Microbiol 20:549±558. 

Pettis GS, Cohen SN 2000): Mutational analysis of the 

tra locus of the broad-host-range Streptomyces plasmid 

pIJ101. J Bacteriol 182:4500± 4504. 

Pittard J, Ramakrishnan T 1964): Gene transfer by F 0 

strains of Escherichia coli: IV. Effect of a chromosomal 

deletion on chromosome transfer. J Bacteriol 

88:367±373. 

Pittard J, Walker EM 1967): Conjugation in Escherichia 

coli: recombination events in terminal regions of 

transferred deoxyribonucleic acid. J Bacteriol 

94:1656±1663. 

Porter RD 1981): Enhanced recombination between 

F42lac and lplac5: Dependence on F42lac fertility 

functions. Mol Gen Genet 184:355±358. 

Porter RD 1982): Recombination properties of P1dlac. 

J Bacteriol 152:345±350. 

Porter RD, McLaughlin T, Low KB 1978): Transduction 

versus ``conjuduction'': Evidence for multiple 

roles for exonuclease V in genetic recombination in 

Escherichia coli. Cold Spring Harbor Symp Quant 

Biol 43:1043±1047. 

Porter RD, Welliver RA, Witkowski TA 1982): Specialized 

transduction with lplac5: Dependence on recB. J 

Bacteriol 150:1485±1488. 

Prangishvili D, Albers S-V, Holz I, Arnold HP, Stedman 

K, Klein T, Singh H, Hiort J, Schweier A, Kristjansson 

JK, Zillig W 1998): Conjugation in Archaea: 

Frequent occurrence of conjugative plasmids in Sulfolobus. 

Plasmid 40:190±202. 

Ramirez-Arcos S, Fernandez-Herrero LA, Marin I, Berenguer 

J 1998): Anaerobic growth, a property horizontally 

transferred by an Hfr-like mechanism among 

extreme thermophiles. J Bacteriol 180:3137±3143. 

Reeves P, Willetts N 1974): Plasmid specificity of the 

origin of transfer of sex factor F. J Bacteriol 120:125± 

130. 

Reygers U, Wessel R, Muller H, Hoffman-Berling H 

1991): Endonuclease activity of Escherichia coli 

DNA helicase I directed against the transfer origin 

of the F factor. EMBO J 10:2689±2694. 

Rice LB 1998): Tn916 family conjugative transposons 

and dissemination of antimicrobial resistance determinants. 

Antimicrob Agents Chemother 42:1871± 

1877. 

Richter A 1957): Complementary determinants on an 

Hfr phenotype in E. coli K12. Genetics 42:391. 

Royle PL, Matsumoto H, Holloway BW 1981): Genetic 

circularity of the Pseudomonas aeruginosa PAO 

chromosome. J Bacteriol 145:145±155. 

Rupp WD, Ihler G 1968): Strand selection during bacterial 

mating. Cold Spring Harbor Symp Quant Biol 

33:647±650. 

Salyers AA, Shoemaker NB, Stevens AM, Li L-Y 

1995): Conjugative transposons: An unusual and diverse 

set of integrated gene transfer elements. Microbiol 

Rev 59:579±590. 

Sancar A, Wharton RP, Seltzer S, Kacinski BM, Clarke 

ND, Rupp WD 1981): Identification of the uvrA gene 

product. J Mol Biol 148:45±62.


Sanderson KE, Ross H, Zeigler L, Makela PH 1972): 

F ‡ , Hfr, and F 0 strains of Salmonella typhimurium and 

Salmonella abony. Bacteriol Rev 36:608±637. 

Sanderson KE, Kadam SK, MacLachlan PR 1983): 

Derepression of F factor function in Salmonella typhimurium. 

Can J Microbiol 29:1205±1212. 

Sanderson KE 1996): F-mediated conjugation, F ‡ 

strains, and Hfr strains of Salmonella typhimurium 

and Salmonella abony. In Neidhardt FC, Curtiss III 

R, Ingraham JL, Lin ECC, Low KB, Magasanik B, 

Reznikoff WS, Riley M, Schaechter M, Umbargar 

HE eds): `Èscherichia coli and Salmonella: Cellular 

and Molecular Biology''. Washington, DC: ASM 

Press, pp 2406±2412. 

Sarathy PV, Siddiqi O 1973): DNA synthesis during 

bacterial conjugation: II. Is DNA replication in the 

Hfr obligatory for chromosome transfer J Mol Biol 

78:443± 451. 

Scaife J, Pekhov AP 1964): Deletion of chromosomal 

markers in association with F-prime factor formation 

in Escherichia coli K12. Genet Res 5:495± 498. 

Schleper C, Holz I, Janekovic D, Murphy J, Zillig W 

1995): A multicopy plasmid of the extremely thermophilic 

archaeon Sulfolobus effects its transfer to recipients 

by mating. J Bacteriol 177:4417± 4426. 

Schmidt KJ, Beck KE, Grogan DW 1999): UV stimulation 

of chromosomal marker exchange in Sulfolobus 

acidocaldarius: Implications for DNA repair, conjugation 

and homologous recombination at extremely 

high temperatures. Genetics 152:1407±1415. 

Scott JR, Kirchman PA, Caparon MG 1988): An intermediate 

in transposition of the conjugative transposon 

Tn916. Proc Natl Acad Sci USA 85:4809±4813. 

Scott JR, Bringel F, Marra D, Van Alstine G, Rudy CK 

1994): Conjugative transposition of Tn916: Preferred 

targets and evidence for conjugative transfer of a 

single strand and for a double-stranded circular intermediate. 


Scott JR, Churchward GG 1995): Conjugative transposition. 

Annu Rev Microbiol 49:367±397. 

Shoemaker NB, Smith MD, Guild WR 1980): DNA seresistant 

transfer of chromosomal cat and tet insertions 

by filter mating in pneumococcus. Plasmid 

3:80±87. 

Seifert HS, Porter RD 1984): Enhanced recombination 

between lplac5 and F42lac: Identification of cis- and 

trans-acting factors. Proc Natl Acad Sci USA 

81:7500±7504. 

Senghas E, Jones JM, Yamamoto M, Gawron-Burke C, 

Clewell DB 1988): Genetic organization of the bacterial 

conjugative transposon Tn916. J Bacteriol 

170:245±249. 

Siddiqi O, Fox MS 1973): Integration of donor DNA in 

bacterial conjugation. J Mol Biol 77:101±123. 

Simons RW, Hughes KT, Nunn WD 1980): Regulation 

of fatty acid degradation in Escherichia coli: Dominance 

studies with strains merodiploid in gene fadR. J 

Bacteriol 143:726±730. 

Smith MD, Shoemaker NB, Burdett V, Guild WR 

1980): Transfer of plasmids by conjugation in 

Streptococcus pneumoniae. Plasmid 3:70±79. 

Smith GR 1991): Conjugational recombination in E. 

coli: Myths and mechanisms. Cell 64:19±27. 

Stedman KM, She Q, Phan H, Holz I, Singh H, Prangishvili 

D, Garrett R, Zillig W 2000): pING family of 

conjugative plasmids from the extremely thermophilic 

archaeon Sulfolobus islandicus: Insights in recombination 

and conjugation in Crenarchaeaota. J Bacteriol 

182:7014 ±7020. 

Stein DS, Kendall KJ, Cohen SN 1989): Identification 

and analysis of transcriptional regulatory signals for 

the kil and kor loci of Streptomyces plasmid pIJ101. 

J Bacteriol 171:5768±5775. 

Thatte V, Iyer VN 1983): Cloning of a plasmid region 

specifying the N transfer system of conjugation in E. 

coli. Gene 21:227±236. 

Thompson TL, Centola MD, Deonier RC 1989): Location 

of the nick at oriT of the F plasmid. J Mol Biol 

207:505±512. 

Timmons MS, Bogardus AM, Deonier RC 1983): Mapping 

of chromosomal IS5 elements that mediate type 

II F-prime plasmid excision in Escherichia coli K-12. J 

Bacteriol 153:395± 407. 

Torres OR, Korman RZ, Zahler SA, Dunny GM 1991): 

The conjugative transposon Tn925: Enhancement of 

conjugal transfer by tetracycline in Enterococcus faecalis 

and mobilization of chromosomal genes in Bacillus 

subtilis and E. faecalis. Mol Gen Genet 

225:395± 400. 

Traxler BA, Minkley EG Jr 1987): Revised genetic map 

of the distal end of the F transfer operon: Implications 

for DNA helicase I, nicking at oriT, and conjugal 

DNA transport. J Bacteriol 169:3251±3259. 

Trieu-Cuot P, Carlier C, Martin P, Courvalin P 1987): 

Plasmid transfer by conjugation from Escherichia coli 

to gram-positive bacteria. FEMS Microbiol Lett 

48:289±294. 

Umeda M, Ohtsubo E 1989): Mapping of insertion 

elements IS1, IS2 and IS3 on the Escherichia coli K- 

12 chromosomes: Role of the insertion elements in 

formation of Hfrs and F 0 factors and in rearrangement 

of bacterial chromosomes. J Mol Biol 208:601± 

614. 

van Biesen T, Frost LS 1994): The FinO protein of IncF 

plasmids binds FinP antisense RNA and its target, 

traJ mRNA, and promotes duplex formation. Mol 

Microbiol 14:427±436. 

Verhoff C, DeHaan PG 1966): Genetic recombination 

in Escherichia coli: I. Relation between linkage of 

unselected markers and map distance. Mutation Res 

3:101±110. 

Vivian A 1971): Genetic control of fertility in Streptomyces 

coelicolor A32): Plasmid involvement in the 

interconversion of UF and IF strains. J Gen Microbiol 

69:353±364.

506PORTER 

Vivian A, Hopwood DA 1973): Genetic control of 

fertility in Streptomyces coelicolor A32): New kinds 

of donor strains. J Gen Microbiol 76:147±162. 

Warren GJ, Twigg AJ, Sherratt DJ 1978): ColE1 plasmid 

mobility and relaxation complex. Nature 

274:259±261. 

Warren GJ, Saul MW, Sherratt DJ 1979): ColE1 plasmid 

mobility: Essential and conditional functions. 


Wilkens B, Lanka E 1993): DNA processing and replication 

during plasmid transfer between Gram-negative 

bacteria. In Clewell DB ed), ``Bacterial Conjugation''. 


Willetts N 1977): The transcriptional control of fertility 

in F-like plasmids. J Mol Biol 112:141±148. 

Willetts N, Maule J 1979): Investigations of the F conjugation 

gene traI: traI mutants and ltraI transducing 

phages. Mol Gen Genet 169:325±336. 

Willetts NS, Skurray R 1980): The conjugative system 

of F-like plasmids. Ann Rev Genet 14:41±76. 

Willetts N, Crowther C 1981): Mobilization of the nonconjugative 

IncQ plasmid RSF1010. Genet Res 

37:311±316. 

Willetts NS, Crowther C, Holloway BW 1981): The 

insertion sequence IS21 of R68.45 and the molecular 

basis for mobilization of the bacterial chromosome. 

Plasmid 6:30±52. 

Willetts N, Skurray R 1987): Structure and function of 

the F factor and mechanism of conjugation. In Neidhardt 

FC, Ingraham JL, Low KB, Magasanik B, 

Schaechter M, Umbargar HE eds), `Èscherichia coli 

and Salmonella typhimurium: Cellular and Molecular 

Biology''. Washington, DC: ASM press, pp 

1110±1133. 

Willetts NS, Wilkins B 1984): Processing of plasmid 

DNA during bacterial conjugation. Microbiol Rev 

48:24± 41. 

Wolk CP, Vonshak A, Kehoe P, Elhai J 1984): Construction 

of shuttle vectors capable of conjugative 

transfer from Escherichia coli to nitrogen fixing filamentous 

cyanobacteria. Proc Natl Acad Sci USA 

81:1561±1565. 

Wollman E-L, Jacob F, Hayes W 1956): Conjugation 

and genetic recombination in Escherichia coli K-12. 

Cold Spring Harbor Symp Quant Biol 21:141±162. 

Wood TH 1968): Effects of temperature, agitation, and 

donor strain on chromosome transfer in Escherichia 

coli K-12. J Bacteriol 96:2077±2084. 

Yancey SD, Porter RD 1985): General recombination 

in Escherichia coli K-12: In vivo role of RecBC 

enzyme. J Bacteriol 162:29±34. 

Yoshioka Y, Ohtsubo H, Ohtsubo E 1987): Repressor 

gene finO in plasmids R100 and F: Constitutive transfer 

of plasmid F is caused by insertion of IS3 into F 

finO. J Bacteriol 169:619±623.

Conjugation

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?